Bug 848684
Summary: | [engine-setup] setup shouldn't fail in case it didn't succeed configuring shared ISO domain (or start NFS service) | ||
---|---|---|---|
Product: | Red Hat Enterprise Virtualization Manager | Reporter: | Haim <hateya> |
Component: | ovirt-engine-setup | Assignee: | Yedidyah Bar David <didi> |
Status: | CLOSED CURRENTRELEASE | QA Contact: | sefi litmanovich <slitmano> |
Severity: | high | Docs Contact: | |
Priority: | medium | ||
Version: | 3.1.0 | CC: | acathrow, alonbl, bazulay, ftaylor, hateya, iheim, jkt, oschreib, pstehlik, Rhev-m-bugs, slitmano, yeylon |
Target Milestone: | --- | ||
Target Release: | 3.3.0 | ||
Hardware: | x86_64 | ||
OS: | Linux | ||
Whiteboard: | integration | ||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2013-09-23 09:19:35 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 1019461 |
Description
Haim
2012-08-16 08:21:36 UTC
(In reply to Haim from comment #0) > Description of problem: > > I don't think that its fair to fail rhevm-setup if shared ISO domain > configuration was not successful. > > we can just alert it as [WARN] and use may configure it later. Any failure from expected result should fail setup. Worse case: run setup again with correct parameters. (In reply to Yedidyah Bar David)
> I tried looking at this today. The root cause, at least in Haim's case, was
> failure of service nfs start. I did not manage to fail it whatever I did,
> except for one thing: it fails if rpcbind is not up. The code (both 3.2 and
> master) first starts rpcbind then nfs. So I guess Forrest's comment about
> "this race condition" refers to rpcbind not really finishing starting before
> nfs tries to connect to it. Not sure how to reproduce this, but we can add
> a loop of checks e.g. running 'rpcinfo -p' before starting nfs, to make sure
> rpcbind is working.
Well, in this case it is not our bug but rhel bug.
As we start a service and we expect it to be active.
I do not mind to add something that retry communicate with nfs...
I mean:
# service nfs start
# <check nfs up in loop>
but if the following fails:
# service rpmbind start
# service nfs start
then it is rhel bug, as same happens at boot.
(In reply to Alon Bar-Lev from comment #5) > (In reply to Yedidyah Bar David) > > I tried looking at this today. The root cause, at least in Haim's case, was > > failure of service nfs start. I did not manage to fail it whatever I did, > > except for one thing: it fails if rpcbind is not up. The code (both 3.2 and > > master) first starts rpcbind then nfs. So I guess Forrest's comment about > > "this race condition" refers to rpcbind not really finishing starting before > > nfs tries to connect to it. Not sure how to reproduce this, but we can add > > a loop of checks e.g. running 'rpcinfo -p' before starting nfs, to make sure > > rpcbind is working. > > Well, in this case it is not our bug but rhel bug. I agree, but am not sure that this was the original problem. If someone can reproduce, it will help. Do remember that the setup code was completely rewritten for 3.3. Haim, can we reproduce this issue? As Didi noted, this code was completely changed in 3.3 the new engine-setup is working fine on 3.3. wasn't able to reproduce the bug. Very well. I am closing this for now. Thanks. |