replication + pruner + watchdog: adjust timeouts based on practical experience

This commit is contained in:
Christian Schwarz 2018-10-21 17:40:46 +02:00
parent b2844569c8
commit 94427d334b
3 changed files with 3 additions and 3 deletions

View File

@ -289,7 +289,7 @@ func (j *ActiveSide) do(ctx context.Context) {
// and the sequential code above transitions to the next state.
go func() {
wdto := envconst.Duration("ZREPL_JOB_WATCHDOG_TIMEOUT", 1*time.Minute)
wdto := envconst.Duration("ZREPL_JOB_WATCHDOG_TIMEOUT", 10*time.Minute)
jitter := envconst.Duration("ZREPL_JOB_WATCHDOG_JITTER", 1*time.Second)
// shadowing!
log := log.WithField("watchdog_timeout", wdto.String())

View File

@ -117,7 +117,7 @@ func NewPrunerFactory(in config.PruningSenderReceiver, promPruneSecs *prometheus
f := &PrunerFactory{
senderRules: keepRulesSender,
receiverRules: keepRulesReceiver,
retryWait: envconst.Duration("ZREPL_PRUNER_RETRY_INTERVAL", 4 * time.Second),
retryWait: envconst.Duration("ZREPL_PRUNER_RETRY_INTERVAL", 10 * time.Second),
considerSnapAtCursorReplicated: considerSnapAtCursorReplicated,
promPruneSecs: promPruneSecs,
}

View File

@ -194,7 +194,7 @@ func resolveConflict(conflict error) (path []*pdu.FilesystemVersion, msg string)
return nil, "no automated way to handle conflict type"
}
var RetryInterval = envconst.Duration("ZREPL_REPLICATION_RETRY_INTERVAL", 4 * time.Second)
var RetryInterval = envconst.Duration("ZREPL_REPLICATION_RETRY_INTERVAL", 10 * time.Second)
type Error interface {
error