Commit Graph

6 Commits

Author SHA1 Message Date
Christian Schwarz
af2d6579c5 [#347] zfscmd: fix dangling trace Task on .Start() failure
fixes #347
2020-09-02 22:45:44 +02:00
Christian Schwarz
0f3da73ef1 [#347] zfscmd + zfs: define .Start() semantics, apply to call sites in pkg zfs
fixes #347
2020-09-02 22:45:44 +02:00
Christian Schwarz
10a14a8c50 [#307] add package trace, integrate it with logging, and adopt it throughout zrepl
package trace:

- introduce the concept of tasks and spans, tracked as linked list within ctx
    - see package-level docs for an overview of the concepts
    - **main feature 1**: unique stack of task and span IDs
        - makes it easy to follow a series of log entries in concurrent code
    - **main feature 2**: ability to produce a chrome://tracing-compatible trace file
        - either via an env variable or a `zrepl pprof` subcommand
        - this is not a CPU profile, we already have go pprof for that
        - but it is very useful to visually inspect where the
          replication / snapshotter / pruner spends its time
          ( fixes #307 )

usage in package daemon/logging:

- goal: every log entry should have a trace field with the ID stack from package trace

- make `logging.GetLogger(ctx, Subsys)` the authoritative `logger.Logger` factory function
    - the context carries a linked list of injected fields which
      `logging.GetLogger` adds to the logger it returns
    - `logging.GetLogger` also uses package `trace` to get the
      task-and-span-stack and injects it into the returned logger's fields
2020-05-19 11:30:02 +02:00
Christian Schwarz
aed6149c8c zfscmd: fix crash in zfscmd_prometheus.go due to incorrectly extracted ProcessState
fixup of 96e188d7c4
refs #196
refs #301

panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x10 pc=0x9a472a]

goroutine 15826 [running]:
os.(*ProcessState).systemTime(...)
        /home/cs/go1.13/src/os/exec_unix.go:98
os.(*ProcessState).SystemTime(...)
        /home/cs/go1.13/src/os/exec.go:141
github.com/zrepl/zrepl/zfs/zfscmd.waitPostPrometheus(0xc000c04800, 0xe21ce0, 0xc000068270, 0xbf9f80d88107e861, 0x19bae710e6, 0x13a8b60)
        /home/cs/zrepl/zrepl/zfs/zfscmd/zfscmd_prometheus.go:69 +0x22a
github.com/zrepl/zrepl/zfs/zfscmd.(*Cmd).waitPost(0xc000c04800, 0xe21ce0, 0xc000068270)
        /home/cs/zrepl/zrepl/zfs/zfscmd/zfscmd.go:155 +0x18a
github.com/zrepl/zrepl/zfs/zfscmd.(*Cmd).CombinedOutput(0xc000c04800, 0xc0004b8270, 0xd02eea, 0x3, 0xc0001f6c40, 0x3)
        /home/cs/zrepl/zrepl/zfs/zfscmd/zfscmd.go:40 +0xb3
github.com/zrepl/zrepl/zfs.ZFSRelease(0xe36aa0, 0xc0004b8270, 0xc0009a3a40, 0x13, 0xc0004a5d00, 0x1, 0x1, 0xed62eb221, 0x13a8b60)
        /home/cs/zrepl/zrepl/zfs/holds.go:102 +0x2a7
github.com/zrepl/zrepl/endpoint.ReleaseStep(0xe36aa0, 0xc0004b8270, 0xc0004befc0, 0xe, 0xd08482, 0x8, 0xc0001cb02f, 0x2, 0x1eeea3bff89dc90b, 0x134d6, ...)
        /home/cs/zrepl/zrepl/endpoint/endpoint_zfs_abstraction_step.go:130 +0x367
github.com/zrepl/zrepl/endpoint.(*Sender).SendCompleted.func2(0xc000459190, 0xc000390e30, 0xc00041fd80, 0xc0004befc0, 0xe, 0xd08482, 0x8, 0xc0001cb02f, 0x2, 0x1eeea3bff89dc90b, ...)
        /home/cs/zrepl/zrepl/endpoint/endpoint.go:419 +0x1c3
created by github.com/zrepl/zrepl/endpoint.(*Sender).SendCompleted
        /home/cs/zrepl/zrepl/endpoint/endpoint.go:413 +0x776
2020-04-21 14:10:25 +02:00
Christian Schwarz
0834a184b8 zfscmd: do not do duplicate waitPre callbacks
it just makes sense that if we only dispatch one waitPost, we should
also only dispatch one waitPre
2020-04-21 14:10:18 +02:00
Christian Schwarz
1336c91865 zfs: introduce pkg zfs/zfscmd for command logging, status, prometheus metrics
refs #196
2020-04-05 20:47:25 +02:00