#2626 Retry EPIPE from clients
Closed: Fixed None Opened 4 years ago by jhrozek.

Simo said:

We should handle EPIPE, if sssd_nss crashes and is restarted while a client is trying to send info we'd probably get EPIPE.

While thinking about that I think we should retry also if we get EPIPE on recv(), but in that case we cannot just retry the recv().
In gneral when we get EPIPE we do not know what the server recieved so we should retry the whole operation.

If we add EPIPE handling though the number of retries should be limited though, to avoid infinite loops (both on send() and recv() errors).


Fields changed

blockedby: =>
blocking: =>
changelog: =>
coverity: =>
design: =>
design_review: => 0
feature_milestone: =>
fedora_test_page: =>
mark: no => 0
milestone: NEEDS_TRIAGE => SSSD 1.14 beta
review: True => 0
selected: =>
testsupdated: => 0

Moving up to needs_triage, we might need to do this sooner.

milestone: SSSD 1.14 beta => NEEDS_TRIAGE
sensitive: => 0

This ticket was previously in 1.14, but since downstream needs this fix sooner, we should add it to 1.13.3

Fields changed

milestone: NEEDS_TRIAGE => SSSD 1.13.3
owner: somebody => lslebodn

Lukas is still looking into the issue, moving to 1.13.4

milestone: SSSD 1.13.3 => SSSD 1.13.4

Fields changed

patch: 0 => 1
status: new => assigned

master:

sssd-1-13:

resolution: => fixed
status: assigned => closed

Metadata Update from @jhrozek:
- Issue assigned to lslebodn
- Issue set to the milestone: SSSD 1.13.4

2 years ago

Login to comment on this ticket.

Metadata