30 minutes sounds like a TCP timeout. Are there any firewalls or NAT 
devices in the path? Sometimes they close the connection when the first FIN 
is sent but before the last FIN-ACK and that can cause SSH to hang up. 

A second possibility is asymetric routes where the replies come in on a 
different interface then the outgoing packets.

In either case, tcpdump can pinpoint the issue. 



On Monday, December 20, 2021 at 4:27:50 PM UTC-5 [email protected] 
wrote:

>
> We use Jenkins to check that various things are running on our 
> infrastructure hosts.
> This is typically done with something like:
>
>      ssh hostname /the/script/to/check
>
> If the check exits with a zero status, Jenkins goes on to mark the build 
> as good.
>
> Sometime in the last couple months, the script and the ssh exits with a 
> zero status
> but Jenkins hangs on and does not mark the job as complete for a long 
> time, like close to 30 minutes.
>
> If we invoke these manually, there is no issue.   The check runs very 
> quickly (under a couple seconds).
>
> We've updated Jenkins several times (now on 2.325) in an attempt to fix 
> the issue.
>
> Any ideas where to look?
>
> Thanks in advance.
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/d6b13386-9a91-4b4c-9764-ddea9323048dn%40googlegroups.com.

Reply via email to