On 2026-03-03 09:24, Fabian Grünbichler wrote:
> On March 3, 2026 8:15 am, Hannes Laimer wrote:
>> If the worker finishes right after we `waitpid` but before we add it to
>> `WORKER_PIDS` the `worker_reaper` won't `waitpid` it cause it iterates
>> over `WORKER_PIDS`. So
> 
> it would be interesting to get more details how this happens in practice
> (with your reproducer)?
> 

I do have a reproducer for task processes sticking around as zombies
when they are done, but this change unfortunately did not fix that. I
just noticed this in the process of finding the cause for the "original"
problem, so I guess this is not a problem in practice, cause of the
tight timings? But technically it would be possible (I think)

> the sequence when forking a worker is:
> - fork
> - child executes some setup code
> - child tells parent it is ready
> - child waits for parent to tell it it can continue
> 
> register_worker is called by the parent in between the last two steps
> (after receiving the notification form the child, but before sending the
> notification to the child), so why does the child disappear inbetween?
> 
> I think this might actually (also?) be missing error handling in
> fork_worker? all the POSIX::close/read/write calls there don't check for
> failure, which means we attempt to register a worker that has already
> failed at that point?
> 

could be, but I don't think that should influence if a `SIGCHLD` is sent
when the child is done? Cause the handler for `SIGCHLD` in the parent is
never called...
I'll take a look at that, thanks for the pointer!

> and, somewhat tangentially related - should we switch this code over to
> use pidfds and waitid to close PID reuse races?
> 

@Wolfgang also mentioned that, would probably make sense

>>  - the clean-up triggered by the SIGCHLD won't catch it cause it needs it to
>>    be in `WORKER_PIDS`
>>  - and, `register_worker` won't because it was still running when it
>>    `waitpid`'ed it
>>
>> Moving the insertion into `WORKER_PIDS` before the `waitpid` solves
>> this by making sure it is
>>  - always in the var for `worker_reaper`
>>  - and, if SIGCHILD should trigger `worker_reaper` before we add it to
>>    `WORKER_PIDS`, the `waitpid` in `register_worker` itself will catch
>>    it
>>
>> Signed-off-by: Hannes Laimer <[email protected]>
>> ---
>>  src/PVE/RESTEnvironment.pm | 11 ++++++-----
>>  1 file changed, 6 insertions(+), 5 deletions(-)
>>
>> diff --git a/src/PVE/RESTEnvironment.pm b/src/PVE/RESTEnvironment.pm
>> index 4ed5c05..4677687 100644
>> --- a/src/PVE/RESTEnvironment.pm
>> +++ b/src/PVE/RESTEnvironment.pm
>> @@ -99,17 +99,18 @@ my $register_worker = sub {
>>  
>>      return if !$pid;
>>  
>> -    # do not register if already finished
>> +    $WORKER_PIDS->{$pid} = {
>> +        user => $user,
>> +        upid => $upid,
>> +    };
>> +
>> +    # remove immediately if already finished
>>      my $waitpid = waitpid($pid, WNOHANG);
>>      if (defined($waitpid) && ($waitpid == $pid)) {
>>          delete($WORKER_PIDS->{$pid});
>>          return;
>>      }
>>  
>> -    $WORKER_PIDS->{$pid} = {
>> -        user => $user,
>> -        upid => $upid,
>> -    };
>>  };
>>  
>>  # initialize environment - must be called once at program startup
>> -- 
>> 2.47.3
>>
>>
>>
>>
>>
>>




Reply via email to