[SOLVED] ZCO 6.0.4 inifinite loop

[Rich Graves]

For over a week I've had a ZCO freaking out trying and failing to do the same thing with the same items, up to hundreds of times per second. (How) can I convince ZCO to give up?

2009-12-29 10:46:10,273 INFO  [btpool0-14067://mail.example.com/service/soap/MsgActionRequest] [name=user@example.com;mid=3324;ip=192.168.192.168;ua=ZimbraConnectorForOutlook/6.0.5797.4;] SoapEngine - handler exception
com.zimbra.cs.mailbox.MailServiceException$NoSuchItemException: no such message: 15554
ExceptionId:btpool0-14067://mail.example.com/service/soap/MsgActionRequest:1262105170273:abe4414501ebba8c
--
2009-12-29 10:46:10,280 INFO  [btpool0-14064://mail.example.com/service/soap/MsgActionRequest] [name=user@example.com;mid=3324;ip=192.168.192.168;ua=ZimbraConnectorForOutlook/6.0.5797.4;] SoapEngine - handler exception
com.zimbra.cs.mailbox.MailServiceException$NoSuchItemException: no such message: 15563
ExceptionId:btpool0-14064://mail.example.com/service/soap/MsgActionRequest:1262105170279:abe4414501ebba8c
--
2009-12-29 10:46:10,286 INFO  [btpool0-14064://mail.example.com/service/soap/MsgActionRequest] [name=user@example.com;mid=3324;ip=192.168.192.168;ua=ZimbraConnectorForOutlook/6.0.5797.4;] SoapEngine - handler exception
com.zimbra.cs.mailbox.MailServiceException$NoSuchItemException: no such message: 15554
ExceptionId:btpool0-14064://mail.example.com/service/soap/MsgActionRequest:1262105170285:abe4414501ebba8c
--
2009-12-29 10:46:10,292 INFO  [btpool0-14064://mail.example.com/service/soap/MsgActionRequest] [name=user@example.com;mid=3324;ip=192.168.192.168;ua=ZimbraConnectorForOutlook/6.0.5797.4;] SoapEngine - handler exception
com.zimbra.cs.mailbox.MailServiceException$NoSuchItemException: no such message: 15563
ExceptionId:btpool0-14064://mail.example.com/service/soap/MsgActionRequest:1262105170291:abe4414501ebba8c
--
2009-12-29 10:46:10,297 INFO  [btpool0-14064://mail.example.com/service/soap/MsgActionRequest] [name=user@example.com;mid=3324;ip=192.168.192.168;ua=ZimbraConnectorForOutlook/6.0.5797.4;] SoapEngine - handler exception
com.zimbra.cs.mailbox.MailServiceException$NoSuchItemException: no such message: 15554
ExceptionId:btpool0-14064://mail.example.com/service/soap/MsgActionRequest:1262105170297:abe4414501ebba8c
--
2009-12-29 10:46:10,304 INFO  [btpool0-14064://mail.example.com/service/soap/MsgActionRequest] [name=user@example.com;mid=3324;ip=192.168.192.168;ua=ZimbraConnectorForOutlook/6.0.5797.4;] SoapEngine - handler exception
com.zimbra.cs.mailbox.MailServiceException$NoSuchItemException: no such message: 15563
ExceptionId:btpool0-14064://mail.example.com/service/soap/MsgActionRequest:1262105170303:abe4414501ebba8c

[fsiegel]

Bug 43432 and 43433, fixed in 6.0.5. Not sure if there is a way to stop this besides getting the new version, or going back a version. The bug was introduced in 6.0.4 -- previous versions do not have it. The problem is we're retrying NO_SUCH_MSG when we shouldn't. Deleting the change record could help, but you would need help to do that.

I don't know how you got in the no such msg situation. Did you delete something on the server, and also change it in Outlook while offline?

A new profile would probably help, but of course that's a pain.

[Rich Graves]

How about we downgrade to 6.0.3, sync, and then re-upgrade to 6.0.4?

Deleting the change record could help, but you would need help to do that.

Help gratefully accepted. :-)

[fsiegel]

Actually, that's not a bad idea. That should delete the change record. As long as you don't get in the NO_SUCH_MSG situation again, that could work. Definitely worth a try. To downgrade, you'll have to uninstall and reinstall zco, but that's no big deal.

[Rich Graves]

Done -- roll back to 6.0.3, then forward to 6.0.4 (we need at least 15554
and 41294 fixed). Seems happy.

I tried the Zimbra Support toolbar to open by zid, and Outlook didn't know of those IDs, either.

It's possible that grepping the logs might turn up some message that was created and quickly deleted... but the nature of the bug is that my logs got really, really big. Calling it resolved, hoping not to see the bug again until 6.0.5 (or a hotfix) is available.

[swix]

Same problem here since a few days on a quite "new" 6.0.4 setup (so no rollback possible). So I guess my only option is to upgrade to 6.0.5... ? Or is there any other workaround I can try? (my mailbox.log is growing quite fast, about 10-20GB per day!).

[fsiegel]

Upgrading to 6.0.5 is the way to go. In this case, you can just try upgrading the Outlook connector -- you don't have to go through the whole server upgrade. Just install the 6.0.5 ZCO .msi right over the top, and you should be fine.

Or I suppose you could create a new profile, and initial sync -- that might solve it for awhile, but upgrading is the better option.

[burnmage]

Bug 43432 and 43433, fixed in 6.0.5. Not sure if there is a way to stop this besides getting the new version, or going back a version. The bug was introduced in 6.0.4 -- previous versions do not have it.

I am running:

Release 5.0.21_GA_3150.RHEL5_64_20091211200605 RHEL5_64 NETWORK edition

RHEL is patched 'pretty' current (couple updates I'm installing right now)

logs were eating 140GB of space from this until I discovered it and started cleaning up.

[fsiegel]

Well if you're running a 5.0.21 ZCS server, you shouldn't be running a 6.0.x ZCO. In this case, the bug was fixed in 5.0.22, so you could upgrade ZCO to 5.0.22 or 5.0.23. FWIW, stay away from ZCO 5.0.20 and 5.0.21.

[burnmage]

Sorry, I forgot about the ZCO 6.0.x part, but no I was running a 5.0.21 ZCO with the same ZCS. I've installed the ZCO 5.0.22 connector on the system that was generating the errors and should be upgrading ZCS tonight to the same.