🐛 Prevent race when informers are started more than once #2758
Conversation
k8s-ci-robot
added
the
cncf-cla: yes
|
Done |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: alvaroaleman The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
added
approved
|
/hold |
k8s-ci-robot
added
the
do-not-merge/hold
pkg/cache/internal/informers.go
Outdated
| @@ -186,6 +187,12 @@ type Informers struct { | |||
| // Start calls Run on each of the informers and sets started to true. Blocks on the context. | |||
| // It doesn't return start because it can't return an error, and it's not a runnable directly. | |||
| func (ip *Informers) Start(ctx context.Context) error { | |||
| select { | |||
| case <-ip.startWait: | |||
There was a problem hiding this comment.
Do we need a lock around this?
There was a problem hiding this comment.
No, this is a channel
There was a problem hiding this comment.
That's not how I meant it :)
What if you have one goroutine in l.198 and another one then just going through l.191?
I would assume the same data race.
Don't we have to make sure an additional goroutine can't go through l.190-l.195 while another one is inside
l.197-216?
There was a problem hiding this comment.
Put all of it under the lock
k8s-ci-robot
added
size/S
|
/hold cancel |
k8s-ci-robot
removed
the
do-not-merge/hold
alvaroaleman
force-pushed
the
fix-race
branch
2 times, most recently
from
b9c9be8
to
c2c96b0
Compare
k8s-ci-robot
added
size/M
If `Informers` are started a second time, there is a possibility for a data race because it sets a `ctx` field on itself. This write is protected by a mutex, but reads from that field are not.
|
/lgtm /hold |
k8s-ci-robot
added
the
do-not-merge/hold
k8s-ci-robot
added
the
lgtm
|
LGTM label has been added. Git tree hash: 59bffd29e2eca09280185b4878884fe155112efb
|
|
/hold cancel |
k8s-ci-robot
removed
the
do-not-merge/hold
If
Informersare started a second time, there is a possibility for a data race because it sets actxfield on itself. This write is protected by a mutex, but reads from that field are not.As it is generally unclear and untested what happens when a cache is started a second time, simply error out to make the user aware that they were starting the cache a second time.