How is upgrade from 11.1 to 11.2?

[techsharp]

jaime -

I just tried an old email with 11.1 and put in 11.2 and I put the username/password and it worked -

interesting how it did not for you? Try it again just to make sure..

[Kris]

Well.. it's been a busy day for me.

Started at 8:00 at my office.. making backups of the old Scalix 10 server.. replaced the OS disks with new ones.. Installed RHEL 5 (old server was using RHEL3, and I couldn't use that with Scalix 11). Installed Scalix 11. So far, everything looks good. Had some little problems, but not really Scalix related problems.

At the moment, sxmkindex is running (300GB mailstore is being indexed.. ). Worked until 18:15 and went home. At home, installed the new Scalix connector on my home computer with smart cache enabled. Did the first sync through VPN.. took a while, but now it's working great..

I'm also very happy and impressed with the search.. works great! Tomorrow going back to the office to do some other updates.. and then I'll wait until monday.. hopefully there won't be any problems with the users..

But so far... VERY positive about Scalix 11.2.!!

[chris]

@ALL: I split the discussions around postgres off into a seperate topic so others can find the information more easily.

The thread is located at viewtopic.php?t=8917

Chris

[jhamill]

jaime -

I just tried an old email with 11.1 and put in 11.2 and I put the username/password and it worked -
interesting how it did not for you? Try it again just to make sure..

that also worked flawlessly for me - as did the upgrade.

Also if anyone is interested we also migrated our scalix system onto an iScsi volume over the weekend rather than using the raided sata drives in the local server. the disktest benchmark indicated the iscsi volume would perform better than the local disks and for the past few days it appears to be working very well.

[adhodgson]

Hi,

Just to add to the good comments. I upgraded from 11.1 to 11.2 on Monday evening, it took less than 10 minutes. I also did an autoupgrade of the connecter on all workstations, and users can now do the shared calendars properly - multiple shared calendars, no more free/busy error messages!

I did a full omscan with fix mode before upgrading during the day just to check.

Andrew.

[LHD-Tech]

Hey guys,
I am glad that everyone seems to be having no troubles upgrading. Makes me very confident that I will have like results. The only thing that is holding me back from upgrading from 10.x to 11.2 is that I dont see much of a need in my company. We have about 150 premium users.
- Scalix mobile looked like a good idea, but its only your inbox. Had it been a calendar too, it would have been awesome! Our phones are pocket PCs so we can get to our /webmail page already.
- Smartcache sounds like an awesome idea and I am glad that it would speed things up a bit but i have no complaints of speed being an issue.
- I am happy to see sxaa is included with the install. But not a big need for that.
- I like the import/export tools, but i dont think it would be used often.

I am guessing it would be best to upgrade to 11.2 but I dont want to fix something that is not broke. Am i missing something as to why i should upgrade sooner than later?

Did Scalix Connect for Outlook have to be uninstalled and reinstalled on each workstation?

[obrodkin]

I am holding back on upgrading (11.0.2), after reading couple of horror stories on this board. Hopefully this is not a case for 11.2.

So... I understand most of you are upgrading the existing server rather than doing a clean install and exporting/importing data storage (/var/opt/scalix). In my case, upgrading production server is ... unnerving. I rather spend time dealing with data migration while the company's email is up, than troubleshooting what went wrong during the upgrade and having people breathing to my neck.

[pcrock]

So far so good, ish, since the upgrade I'm getting issues with both sac and webmail dying after being up for a while with this error in catalina.out

Exception in thread "QueueManager" java.lang.OutOfMemoryError: Java heap space

The page occasionally says 'the server didn't return a valid XML' or 'Java Heap Space'

Stop and restart them and it's ok for a bit then dies again. I never saw this with the older releases.

Other than that, the pda users are very happy bunnies now.

[konman]

Guys - could someone give a quick and dirty steps for going from 11.1->11.2?
Which commands to stop the relevant services before the upgrade?

Thanks,
KonMan.

[kfleming]

The install overwrote my smtpd.cfg file, meaning that spamassassin wasn't being called, and it added a new file in /etc/httpd/conf.d, which broke the websites also hosted on the server. Luckily I had all this backed up, otherwise I would have been in trouble!

[kanderson]

Konman

omshut -t 3
/etc/init.d/scalix-tomcat stop
/etc/init.d/scalix-postgres stop
run the installer. Either command line or graphically.

You're done. The services will all be started automatically at the end of the upgrade process.

Kev.

[konman]

Konman

omshut -t 3
/etc/init.d/scalix-tomcat stop
/etc/init.d/scalix-postgres stop
run the installer. Either command line or graphically.

You're done. The services will all be started automatically at the end of the upgrade process.

Kev.

Kev - Many thanks..
KonMan

[jaime.pinto]

Do a reboot of the server at the end, just to be sure everything comes back as expected: sac, SWA, https redirection, (anti-spam, anti-virus if it's the case)

Also omstat -a

[scalixuser]

I untarred, started ./scalix-installer (on CentOS 4.5; 11.1.0->11.2), but did not do the following before running the installer:
omshut -t 3
/etc/init.d/scalix-tomcat stop
/etc/init.d/scalix-postgres stop

That killed the upgrade (error message: scalix The requested resource (/webmail/) is not available.)
Luckily there was an easy fix: just uninstall, reinstall the new Scalix.
It would be nice for people like me who do not read READMEs (guilty as charged), if the install script would take care or stopping scalix-tomcat and scalix-postgres.

Otherwise very happy with this upgrade!