rclone/vfs/vfscache/cache_test.go

730 lines
18 KiB
Go
Raw Normal View History

package vfscache
import (
"context"
"fmt"
"os"
"path/filepath"
"sort"
"testing"
"time"
_ "github.com/rclone/rclone/backend/local" // import the local backend
"github.com/rclone/rclone/fs"
"github.com/rclone/rclone/fs/config"
"github.com/rclone/rclone/fstest"
"github.com/rclone/rclone/lib/diskusage"
"github.com/rclone/rclone/vfs/vfscommon"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// TestMain drives the tests
func TestMain(m *testing.M) {
fstest.TestMain(m)
}
// convert c.item to a string
func itemAsString(c *Cache) []string {
c.mu.Lock()
defer c.mu.Unlock()
var out []string
for name, item := range c.item {
out = append(out, fmt.Sprintf("name=%q opens=%d size=%d", filepath.ToSlash(name), item.opens, item.info.Size))
}
sort.Strings(out)
return out
}
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
// convert c.item to a string
func itemSpaceAsString(c *Cache) []string {
c.mu.Lock()
defer c.mu.Unlock()
var out []string
for name, item := range c.item {
space := item.info.Rs.Size()
out = append(out, fmt.Sprintf("name=%q opens=%d size=%d space=%d", filepath.ToSlash(name), item.opens, item.info.Size, space))
}
sort.Strings(out)
return out
}
// open an item and write to it
func itemWrite(t *testing.T, item *Item, contents string) {
require.NoError(t, item.Open(nil))
_, err := item.WriteAt([]byte(contents), 0)
require.NoError(t, err)
}
func assertPathNotExist(t *testing.T, path string) {
_, err := os.Stat(path)
assert.True(t, os.IsNotExist(err))
}
func assertPathExist(t *testing.T, path string) os.FileInfo {
fi, err := os.Stat(path)
assert.NoError(t, err)
return fi
}
type avInfo struct {
Remote string
Size int64
IsDir bool
}
var avInfos []avInfo
func addVirtual(remote string, size int64, isDir bool) error {
avInfos = append(avInfos, avInfo{
Remote: remote,
Size: size,
IsDir: isDir,
})
return nil
}
func newTestCacheOpt(t *testing.T, opt vfscommon.Options) (r *fstest.Run, c *Cache) {
r = fstest.NewRun(t)
ctx, cancel := context.WithCancel(context.Background())
avInfos = nil
c, err := New(ctx, r.Fremote, &opt, addVirtual)
require.NoError(t, err)
t.Cleanup(func() {
err := c.CleanUp()
require.NoError(t, err)
assertPathNotExist(t, c.root)
cancel()
})
return r, c
}
func newTestCache(t *testing.T) (r *fstest.Run, c *Cache) {
opt := vfscommon.Opt
// Disable the cache cleaner as it interferes with these tests
opt.CachePollInterval = 0
// Disable synchronous write
opt.WriteBack = 0
return newTestCacheOpt(t, opt)
}
func TestCacheNew(t *testing.T) {
r, c := newTestCache(t)
assert.Contains(t, c.root, "vfs")
assert.Contains(t, c.fcache.Root(), filepath.Base(r.Fremote.Root()))
assert.Equal(t, []string(nil), itemAsString(c))
// createItemDir
p, err := c.createItemDir("potato")
require.NoError(t, err)
assert.Equal(t, "potato", filepath.Base(p))
assert.Equal(t, []string(nil), itemAsString(c))
fi := assertPathExist(t, filepath.Dir(p))
assert.True(t, fi.IsDir())
// get
item, _ := c.get("potato")
item2, _ := c.get("potato")
assert.Equal(t, item, item2)
assert.WithinDuration(t, time.Now(), item.info.ATime, time.Second)
// open
assert.Equal(t, []string{
`name="potato" opens=0 size=0`,
}, itemAsString(c))
potato := c.Item("/potato")
require.NoError(t, potato.Open(nil))
assert.Equal(t, []string{
`name="potato" opens=1 size=0`,
}, itemAsString(c))
assert.WithinDuration(t, time.Now(), potato.info.ATime, time.Second)
assert.Equal(t, 1, potato.opens)
// write the file
require.NoError(t, potato.Truncate(5))
atime := time.Now()
potato.info.ATime = atime
assert.Equal(t, []string{
`name="potato" opens=1 size=5`,
}, itemAsString(c))
assert.True(t, atime.Equal(potato.info.ATime), fmt.Sprintf("%v != %v", atime, potato.info.ATime))
// try purging with file open
c.purgeOld(10 * time.Second)
assertPathExist(t, p)
// close
assert.Equal(t, []string{
`name="potato" opens=1 size=5`,
}, itemAsString(c))
require.NoError(t, potato.Truncate(6))
assert.Equal(t, []string{
`name="potato" opens=1 size=6`,
}, itemAsString(c))
require.NoError(t, potato.Close(nil))
assert.Equal(t, []string{
`name="potato" opens=0 size=6`,
}, itemAsString(c))
item, _ = c.get("potato")
assert.WithinDuration(t, time.Now(), item.info.ATime, time.Second)
assert.Equal(t, 0, item.opens)
// try purging with file closed
c.purgeOld(10 * time.Second)
assertPathExist(t, p)
//.. purge again with -ve age
c.purgeOld(-10 * time.Second)
assertPathNotExist(t, p)
// clean - have tested the internals already
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.clean(false)
}
func TestCacheOpens(t *testing.T) {
_, c := newTestCache(t)
assert.Equal(t, []string(nil), itemAsString(c))
potato := c.Item("potato")
require.NoError(t, potato.Open(nil))
assert.Equal(t, []string{
`name="potato" opens=1 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Open(nil))
assert.Equal(t, []string{
`name="potato" opens=2 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Close(nil))
assert.Equal(t, []string{
`name="potato" opens=1 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Close(nil))
assert.Equal(t, []string{
`name="potato" opens=0 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Open(nil))
a1 := c.Item("a//b/c/d/one")
a2 := c.Item("a/b/c/d/e/two")
a3 := c.Item("a/b/c/d/e/f/three")
require.NoError(t, a1.Open(nil))
require.NoError(t, a2.Open(nil))
require.NoError(t, a3.Open(nil))
assert.Equal(t, []string{
`name="a/b/c/d/e/f/three" opens=1 size=0`,
`name="a/b/c/d/e/two" opens=1 size=0`,
`name="a/b/c/d/one" opens=1 size=0`,
`name="potato" opens=1 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Close(nil))
require.NoError(t, a1.Close(nil))
require.NoError(t, a2.Close(nil))
require.NoError(t, a3.Close(nil))
assert.Equal(t, []string{
`name="a/b/c/d/e/f/three" opens=0 size=0`,
`name="a/b/c/d/e/two" opens=0 size=0`,
`name="a/b/c/d/one" opens=0 size=0`,
`name="potato" opens=0 size=0`,
}, itemAsString(c))
}
// test the open, createItemDir, purge, close, purge sequence
func TestCacheOpenMkdir(t *testing.T) {
_, c := newTestCache(t)
// open
potato := c.Item("sub/potato")
require.NoError(t, potato.Open(nil))
assert.Equal(t, []string{
`name="sub/potato" opens=1 size=0`,
}, itemAsString(c))
// createItemDir
p, err := c.createItemDir("sub/potato")
require.NoError(t, err)
assert.Equal(t, "potato", filepath.Base(p))
assert.Equal(t, []string{
`name="sub/potato" opens=1 size=0`,
}, itemAsString(c))
// test directory exists
fi := assertPathExist(t, filepath.Dir(p))
assert.True(t, fi.IsDir())
// clean the cache
c.purgeOld(-10 * time.Second)
// test directory still exists
fi = assertPathExist(t, filepath.Dir(p))
assert.True(t, fi.IsDir())
// close
require.NoError(t, potato.Close(nil))
assert.Equal(t, []string{
`name="sub/potato" opens=0 size=0`,
}, itemAsString(c))
// clean the cache
c.purgeOld(-10 * time.Second)
c.purgeEmptyDirs("", true)
assert.Equal(t, []string(nil), itemAsString(c))
// test directory does not exist
assertPathNotExist(t, filepath.Dir(p))
}
func TestCachePurgeOld(t *testing.T) {
_, c := newTestCache(t)
// Test funcs
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.purgeOld(-10 * time.Second)
potato2 := c.Item("sub/dir2/potato2")
require.NoError(t, potato2.Open(nil))
potato := c.Item("sub/dir/potato")
require.NoError(t, potato.Open(nil))
require.NoError(t, potato2.Close(nil))
require.NoError(t, potato.Open(nil))
assert.Equal(t, []string{
`name="sub/dir/potato" opens=2 size=0`,
`name="sub/dir2/potato2" opens=0 size=0`,
}, itemAsString(c))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.purgeOld(-10 * time.Second)
assert.Equal(t, []string{
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
`name="sub/dir/potato" opens=2 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Close(nil))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
assert.Equal(t, []string{
`name="sub/dir/potato" opens=1 size=0`,
}, itemAsString(c))
c.purgeOld(-10 * time.Second)
assert.Equal(t, []string{
`name="sub/dir/potato" opens=1 size=0`,
}, itemAsString(c))
require.NoError(t, potato.Close(nil))
assert.Equal(t, []string{
`name="sub/dir/potato" opens=0 size=0`,
}, itemAsString(c))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.purgeOld(10 * time.Second)
assert.Equal(t, []string{
`name="sub/dir/potato" opens=0 size=0`,
}, itemAsString(c))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.purgeOld(-10 * time.Second)
assert.Equal(t, []string(nil), itemAsString(c))
}
func TestCachePurgeOverQuota(t *testing.T) {
_, c := newTestCache(t)
// Test funcs
// Make some test files
potato := c.Item("sub/dir/potato")
itemWrite(t, potato, "hello")
potato2 := c.Item("sub/dir2/potato2")
itemWrite(t, potato2, "hello2")
assert.Equal(t, []string{
`name="sub/dir/potato" opens=1 size=5`,
`name="sub/dir2/potato2" opens=1 size=6`,
}, itemAsString(c))
// Check nothing removed
c.opt.CacheMaxSize = 1
c.purgeOverQuota()
// Close the files
require.NoError(t, potato.Close(nil))
require.NoError(t, potato2.Close(nil))
assert.Equal(t, []string{
`name="sub/dir/potato" opens=0 size=5`,
`name="sub/dir2/potato2" opens=0 size=6`,
}, itemAsString(c))
// Update the stats to read the total size
c.updateUsed()
// make potato2 definitely after potato
t1 := time.Now().Add(10 * time.Second)
potato2.info.ATime = t1
// Check only potato removed to get below quota
c.opt.CacheMaxSize = 10
c.purgeOverQuota()
assert.Equal(t, int64(6), c.used)
assert.Equal(t, []string{
`name="sub/dir2/potato2" opens=0 size=6`,
}, itemAsString(c))
// Put potato back
potato = c.Item("sub/dir/potato")
require.NoError(t, potato.Open(nil))
require.NoError(t, potato.Truncate(5))
require.NoError(t, potato.Close(nil))
// Update the stats to read the total size
c.updateUsed()
assert.Equal(t, []string{
`name="sub/dir/potato" opens=0 size=5`,
`name="sub/dir2/potato2" opens=0 size=6`,
}, itemAsString(c))
// make potato definitely after potato2
t2 := t1.Add(20 * time.Second)
potato.info.ATime = t2
// Check only potato2 removed to get below quota
c.opt.CacheMaxSize = 10
c.purgeOverQuota()
assert.Equal(t, int64(5), c.used)
c.purgeEmptyDirs("", true)
assert.Equal(t, []string{
`name="sub/dir/potato" opens=0 size=5`,
}, itemAsString(c))
// Now purge everything
c.opt.CacheMaxSize = 1
c.purgeOverQuota()
assert.Equal(t, int64(0), c.used)
c.purgeEmptyDirs("", true)
assert.Equal(t, []string(nil), itemAsString(c))
// Check nothing left behind
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.clean(false)
assert.Equal(t, int64(0), c.used)
assert.Equal(t, []string(nil), itemAsString(c))
}
func TestCachePurgeMinFreeSpace(t *testing.T) {
du, err := diskusage.New(config.GetCacheDir())
if err == diskusage.ErrUnsupported {
t.Skip(err)
}
// We've tested the quota mechanism already, so just test the
// min free space quota is working.
_, c := newTestCache(t)
// First set free space quota very small and check it is OK
c.opt.CacheMinFreeSpace = 1
assert.True(t, c.minFreeSpaceQuotaOK())
assert.True(t, c.quotasOK())
// Now set it a bit larger than the current disk available and check it is BAD
c.opt.CacheMinFreeSpace = fs.SizeSuffix(du.Available) + fs.Gibi
assert.False(t, c.minFreeSpaceQuotaOK())
assert.False(t, c.quotasOK())
}
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
// test reset clean files
func TestCachePurgeClean(t *testing.T) {
r, c := newItemTestCache(t)
contents, obj, potato1 := newFile(t, r, c, "existing")
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
_ = contents
// Open the object to create metadata for it
require.NoError(t, potato1.Open(obj))
require.NoError(t, potato1.Open(obj))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
size, err := potato1.GetSize()
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
require.NoError(t, err)
assert.Equal(t, int64(100), size)
// Read something to instantiate the cache file
buf := make([]byte, 10)
_, err = potato1.ReadAt(buf, 10)
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
require.NoError(t, err)
// Test cache file present
_, err = os.Stat(potato1.c.toOSPath(potato1.name))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
require.NoError(t, err)
// Add some potatoes
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
potato2 := c.Item("sub/dir/potato2")
require.NoError(t, potato2.Open(nil))
require.NoError(t, potato2.Truncate(5))
potato3 := c.Item("sub/dir/potato3")
require.NoError(t, potato3.Open(nil))
require.NoError(t, potato3.Truncate(6))
c.updateUsed()
c.opt.CacheMaxSize = 1
c.purgeClean()
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
assert.Equal(t, []string{
`name="existing" opens=2 size=100 space=0`,
`name="sub/dir/potato2" opens=1 size=5 space=5`,
`name="sub/dir/potato3" opens=1 size=6 space=6`,
}, itemSpaceAsString(c))
assert.Equal(t, int64(11), c.used)
require.NoError(t, potato2.Close(nil))
c.opt.CacheMaxSize = 1
c.purgeClean()
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
assert.Equal(t, []string{
`name="existing" opens=2 size=100 space=0`,
`name="sub/dir/potato3" opens=1 size=6 space=6`,
}, itemSpaceAsString(c))
assert.Equal(t, int64(6), c.used)
require.NoError(t, potato1.Close(nil))
require.NoError(t, potato1.Close(nil))
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
require.NoError(t, potato3.Close(nil))
// Remove all files now. The are all not in use.
// purgeClean does not remove empty cache files. purgeOverQuota does.
// So we use purgeOverQuota here for the cleanup.
c.opt.CacheMaxSize = 1
c.purgeOverQuota()
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
c.purgeEmptyDirs("", true)
vfs: support synchronous cache space recovery upon ENOSPC This patch provides the support of synchronous cache space recovery to allow read threads to recover from ENOSPC errors when cache space can be recovered from cache items that are not in use or safe to be reset/emptied . The patch complements the existing cache cleaning process in two ways. Firstly, the existing cache cleaning process is time-driven that runs periodically. The cache space can run out while the cache cleaner thread is still waiting for its next scheduled run. The io threads encountering ENOSPC return an internal error to the applications in this case even when cache space can be recovered to avoid this error. This patch addresses this problem by having the read threads kick the cache cleaner thread in this condition to recover cache space preventing unnecessary ENOSPC errors from being seen by the applications. Secondly, this patch enhances the cache cleaner to support cache item reset. Currently the cache purge process removes cache items that are not in use. This may not be sufficient when the total size of the working set exceeds the cache directory's capacity. Like in the current code, this patch starts the purge process by removing cache files that are not in use. Cache items whose access times are older than vfs-cache-max-age are removed first. After that, other not-in-use items are removed in LRU order until vfs-cache-max-size is reached. If the vfs-cache-max-size (the quota) is still not reached at this time, this patch adds a cache reset step to reset/empty cache files that are still in use but not dirtied. This enables application processes to continue without seeing an error even when the working set depletes the cache space as long as there is not a large write working set hoarding the entire cache space. By design this patch does not add ENOSPC error recovery for write IOs. Rclone does not empty a write cache item until the file data is written back to the backend upon close. Allowing more cache space to be consumed by dirty cache items when the cache space is already running low would increase the risk of exhausting the cache space in a way that the vfs mount becomes unreadable.
2020-08-25 17:20:29 +02:00
assert.Equal(t, []string(nil), itemAsString(c))
}
func TestCacheInUse(t *testing.T) {
_, c := newTestCache(t)
assert.False(t, c.InUse("potato"))
potato := c.Item("potato")
assert.False(t, c.InUse("potato"))
require.NoError(t, potato.Open(nil))
assert.True(t, c.InUse("potato"))
require.NoError(t, potato.Close(nil))
assert.False(t, c.InUse("potato"))
}
func TestCacheDirtyItem(t *testing.T) {
_, c := newTestCache(t)
assert.Nil(t, c.DirtyItem("potato"))
potato := c.Item("potato")
assert.Nil(t, c.DirtyItem("potato"))
require.NoError(t, potato.Open(nil))
require.NoError(t, potato.Truncate(5))
assert.Equal(t, potato, c.DirtyItem("potato"))
require.NoError(t, potato.Close(nil))
assert.Nil(t, c.DirtyItem("potato"))
}
func TestCacheExistsAndRemove(t *testing.T) {
_, c := newTestCache(t)
assert.False(t, c.Exists("potato"))
potato := c.Item("potato")
assert.False(t, c.Exists("potato"))
require.NoError(t, potato.Open(nil))
assert.True(t, c.Exists("potato"))
require.NoError(t, potato.Close(nil))
assert.True(t, c.Exists("potato"))
c.Remove("potato")
assert.False(t, c.Exists("potato"))
}
func TestCacheRename(t *testing.T) {
_, c := newTestCache(t)
// setup
assert.False(t, c.Exists("potato"))
potato := c.Item("potato")
require.NoError(t, potato.Open(nil))
require.NoError(t, potato.Close(nil))
assert.True(t, c.Exists("potato"))
osPath := c.toOSPath("potato")
osPathMeta := c.toOSPathMeta("potato")
assertPathExist(t, osPath)
assertPathExist(t, osPathMeta)
// rename potato -> newPotato
require.NoError(t, c.Rename("potato", "newPotato", nil))
assertPathNotExist(t, osPath)
assertPathNotExist(t, osPathMeta)
assert.False(t, c.Exists("potato"))
osPath = c.toOSPath("newPotato")
osPathMeta = c.toOSPathMeta("newPotato")
assertPathExist(t, osPath)
assertPathExist(t, osPathMeta)
assert.True(t, c.Exists("newPotato"))
// rename newPotato -> sub/newPotato
require.NoError(t, c.Rename("newPotato", "sub/newPotato", nil))
assertPathNotExist(t, osPath)
assertPathNotExist(t, osPathMeta)
assert.False(t, c.Exists("potato"))
osPath = c.toOSPath("sub/newPotato")
osPathMeta = c.toOSPathMeta("sub/newPotato")
assertPathExist(t, osPath)
assertPathExist(t, osPathMeta)
assert.True(t, c.Exists("sub/newPotato"))
// remove
c.Remove("sub/newPotato")
assertPathNotExist(t, osPath)
assertPathNotExist(t, osPathMeta)
assert.False(t, c.Exists("sub/newPotato"))
// nonexistent file - is ignored
assert.NoError(t, c.Rename("nonexist", "nonexist2", nil))
}
func TestCacheCleaner(t *testing.T) {
opt := vfscommon.Opt
opt.CachePollInterval = fs.Duration(10 * time.Millisecond)
opt.CacheMaxAge = fs.Duration(20 * time.Millisecond)
_, c := newTestCacheOpt(t, opt)
time.Sleep(time.Duration(2 * opt.CachePollInterval))
potato := c.Item("potato")
potato2, found := c.get("potato")
assert.Equal(t, fmt.Sprintf("%p", potato), fmt.Sprintf("%p", potato2))
assert.True(t, found)
for i := 0; i < 100; i++ {
time.Sleep(time.Duration(10 * opt.CachePollInterval))
potato2, found = c.get("potato")
if !found {
break
}
}
assert.NotEqual(t, fmt.Sprintf("%p", potato), fmt.Sprintf("%p", potato2))
assert.False(t, found)
}
func TestCacheSetModTime(t *testing.T) {
_, c := newTestCache(t)
t1 := time.Date(2010, 1, 2, 3, 4, 5, 9, time.UTC)
potato := c.Item("potato")
require.NoError(t, potato.Open(nil))
require.NoError(t, potato.Truncate(5))
require.NoError(t, potato.Close(nil))
c.SetModTime("potato", t1)
osPath := potato.c.toOSPath("potato")
fi, err := os.Stat(osPath)
require.NoError(t, err)
fstest.AssertTimeEqualWithPrecision(t, "potato", t1, fi.ModTime(), time.Second)
}
func TestCacheTotaInUse(t *testing.T) {
_, c := newTestCache(t)
assert.Equal(t, int(0), c.TotalInUse())
potato := c.Item("potato")
assert.Equal(t, int(0), c.TotalInUse())
require.NoError(t, potato.Open(nil))
assert.Equal(t, int(1), c.TotalInUse())
require.NoError(t, potato.Truncate(5))
assert.Equal(t, int(1), c.TotalInUse())
potato2 := c.Item("potato2")
assert.Equal(t, int(1), c.TotalInUse())
require.NoError(t, potato2.Open(nil))
assert.Equal(t, int(2), c.TotalInUse())
require.NoError(t, potato2.Close(nil))
assert.Equal(t, int(1), c.TotalInUse())
require.NoError(t, potato.Close(nil))
assert.Equal(t, int(0), c.TotalInUse())
}
func TestCacheDump(t *testing.T) {
_, c := newTestCache(t)
out := (*Cache)(nil).Dump()
assert.Equal(t, "Cache: <nil>\n", out)
out = c.Dump()
assert.Equal(t, "Cache{\n}\n", out)
c.Item("potato")
out = c.Dump()
want := "Cache{\n\t\"potato\": "
assert.Equal(t, want, out[:len(want)])
c.Remove("potato")
out = c.Dump()
assert.Equal(t, "Cache{\n}\n", out)
}
func TestCacheStats(t *testing.T) {
_, c := newTestCache(t)
out := c.Stats()
assert.Equal(t, int64(0), out["bytesUsed"])
assert.Equal(t, 0, out["erroredFiles"])
assert.Equal(t, 0, out["files"])
assert.Equal(t, 0, out["uploadsInProgress"])
assert.Equal(t, 0, out["uploadsQueued"])
}