Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

client: prevent watching stale alloc state #18612

Merged
merged 4 commits into from
Sep 29, 2023
Merged

client: prevent watching stale alloc state #18612

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
eed4248
client: prevent watching stale alloc state
schmichael Sep 28, 2023
dab3732
off by 1
schmichael Sep 28, 2023
891d2c9
use safetimer and log retry interval
schmichael Sep 29, 2023
c7bab36
sneak in fix for other new time.After use
schmichael Sep 29, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
34 changes: 29 additions & 5 deletions client/allocwatcher/alloc_watcher.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -353,9 +353,11 @@ func (p *remotePrevAlloc) Wait(ctx context.Context) error {
req := structs.AllocSpecificRequest{
AllocID: p.prevAllocID,
QueryOptions: structs.QueryOptions{
Region: p.config.Region,
AllowStale: true,
AuthToken: p.config.Node.SecretID,
Region: p.config.Region,
AuthToken: p.config.Node.SecretID,

// Initially get response from leader, then switch to stale
AllowStale: false,
},
}

Expand All @@ -372,15 +374,36 @@ func (p *remotePrevAlloc) Wait(ctx context.Context) error {
resp := structs.SingleAllocResponse{}
err := p.rpc.RPC("Alloc.GetAlloc", &req, &resp)
if err != nil {
p.logger.Error("error querying previous alloc", "error", err)
retry := getRemoteRetryIntv + helper.RandomStagger(getRemoteRetryIntv)
timer, stop := helper.NewSafeTimer(retry)
p.logger.Error("error querying previous alloc", "error", err, "wait", retry)
select {
case <-time.After(retry):
case <-timer.C:
continue
case <-ctx.Done():
stop()
return ctx.Err()
}
}

// Ensure that we didn't receive a stale response
if req.AllowStale && resp.Index < req.MinQueryIndex {
retry := getRemoteRetryIntv + helper.RandomStagger(getRemoteRetryIntv)
timer, stop := helper.NewSafeTimer(retry)
p.logger.Warn("received stale alloc; retrying",
"req_index", req.MinQueryIndex,
"resp_index", resp.Index,
"wait", retry,
)
select {
case <-timer.C:
continue
case <-ctx.Done():
stop()
return ctx.Err()
}
}

if resp.Alloc == nil {
p.logger.Debug("blocking alloc was GC'd")
return nil
Expand All @@ -392,6 +415,7 @@ func (p *remotePrevAlloc) Wait(ctx context.Context) error {

// Update the query index and requery.
if resp.Index > req.MinQueryIndex {
req.AllowStale = true
req.MinQueryIndex = resp.Index
}
}
Expand Down
4 changes: 3 additions & 1 deletion client/client.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2423,16 +2423,18 @@ OUTER:
// Node.GetClientAllocs which returns older results.
if allocsResp.Index <= allocsReq.MinQueryIndex {
retry := c.retryIntv(getAllocRetryIntv)
timer, stop := helper.NewSafeTimer(retry)
c.logger.Warn("failed to retrieve updated allocs; retrying",
"req_index", allocsReq.MinQueryIndex,
"resp_index", allocsResp.Index,
"num_allocs", len(pull),
"wait", retry,
)
select {
case <-time.After(retry):
case <-timer.C:
Comment on lines -2433 to +2434
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry this is an unrelated fix from a linting error I introduced in my previous PR: #18601

continue
case <-c.shutdownCh:
stop()
return
}
}
Expand Down
Loading