Zimbra+Centos 5.3 x86 + xen = 4gb seg fixup errorerrors?

[phoenix]

Originally Posted by mrball

so... no solution? except change my server OS and all my 10 VMs to 64 bit?

That would be correct, the only supported xen platform for Zimbra is 64bit.

[mrball]

it's been awhile, wonder if anyone's found any alternative solutions?

BTW - I've got a test machine running this with the errors, so far it seems to be ok to just ignore the messages... anybody know if it's safe to ignore these?

[rolandbreedveld]

Originally Posted by mrball

BTW - I've got a test machine running this with the errors, so far it seems to be ok to just ignore the messages... anybody know if it's safe to ignore these?

Yes it is, I have them for a while, it will only slow down your vm a bit, because of the system logger.

There is a Zimbra request for CentOS support, hope the will soon do.
recompile the slap-daemon with nosegneg, will solve it, I think.

[mrball]

after a couple days on a dummy server (no mails coming in), it started getting very high loads from zmlogswatchctl processes... guess it can't just be ignored...

[mrball]

In summary, couple of solutions:

1 - switch to 64bit Xen (i.e. re-install 64 bit Centos, switch all your existing VMs to 64 bit...)

2 - migrate to a different solution (like Google Apps)

(I chose #2) Hope this helps!