Server 11.1 won't install

[jillrae]

Upgraded from 11.0.3 to 11.1 running on a SLES 10. All packages installed except the server package.. So ran installer a seond time. Here is install log output from that attempt:

2007-06-20 10:45:52,638 INFO System check start
2007-06-20 10:45:52,749 INFO No command selinuxenabled found on system, SELinux is not installed
2007-06-20 10:45:52,749 INFO Operation passed: Environment check passed.
2007-06-20 10:45:52,848 INFO Operation passed: Filesystem check passed.
2007-06-20 10:45:53,039 INFO Network interface is eth1
2007-06-20 10:45:53,084 INFO Operation passed: Network check passed.
2007-06-20 10:45:54,341 INFO Apache 2.x appears to be installed.
2007-06-20 10:45:54,341 INFO Operation passed: Dependency check passed.
2007-06-20 10:45:54,934 INFO Operation passed: Running services check passed.
2007-06-20 10:45:54,935 INFO System check end
2007-06-20 10:45:54,936 INFO System check result: pass
2007-06-20 10:45:59,096 INFO Found permanent license file: /var/opt/scalix/lx/s/LicenseKeys/permanent.lic
2007-06-20 10:45:59,143 INFO Found permanent license file: /var/opt/scalix/lx/s/LicenseKeys/permanent.lic
2007-06-20 10:45:59,250 INFO determined license validity: True (LVID: 2007-10-26, issue date of server: 2006-12-15)
2007-06-20 10:45:59,257 INFO Install start
2007-06-20 10:46:00,036 INFO Unlinking 'sxadmin' user from Unix user
2007-06-20 10:46:00,188 ERROR Failed to unlink the user: ['ommodu : [OM 8047] Cannot contact Session Monitor Daemon']
2007-06-20 10:46:00,188 ERROR pre-upgrade function remove_sxadmin_user failed to execute successfully
2007-06-20 10:46:00,305 INFO Server is not running
2007-06-20 10:46:00,320 INFO ldapmapper is not running.
2007-06-20 10:46:00,320 INFO Starting installation of RPM files [('/root/scalix-11.1.0-GA/software/scalix_server/scalix-text-extractors-1.0-1.sles10.i586.rpm', 'scalix-text-extractors'), ('/root/scalix-11.1.0-GA/software/scalix_server/scalix-server-11.1.0.10849-1.sles10.i586.rpm', 'scalix-server')]
2007-06-20 10:46:03,905 ERROR Operation failed: Install /bin/rpm -U -v --percent /root/scalix-11.1.0-GA/software/scalix_server/scalix-text-extractors-1.0-1.sles10.i586.rpm /root/scalix-11.1.0-GA/software/scalix_server/scalix-server-11.1.0.10849-1.sles10.i586.rpm: (2) ['Preparing packages for installation...', 'scalix-text-extractors-1.0-1.sles10', 'error: %pre(scalix-server-11.1.0.10849-1.sles10.i586) scriptlet failed, exit status 10', 'error: install: %pre scriptlet failed (2), skipping scalix-server-11.1.0.10849-1.sles10']
2007-06-20 10:46:03,905 CRITICAL Operation failed: Installation of packages failed
2007-06-20 10:46:03,905 ERROR Operation failed: Aborting installation thread.
2007-06-20 10:46:03,905 INFO Install end

Could this be linked to the fact I can't log into SAC before the upgrade?

Thanks

jillrae

[KevinAnderson]

Can you please post the results of "df -h" from your server?

Thanks
Kev.

[jillrae]

linux:~ # df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda3 13G 3.5G 8.6G 29% /
udev 1014M 156K 1014M 1% /dev
/dev/sda1 2.1G 40M 2.0G 2% /boot
/dev/mapper/vgscalix-lvscalix
28G 9.8G 19G 35% /var/opt/scalix


Here it is.

Thanks

[KevinAnderson]

Shoot, I was expecting that this would be an error because there was a subdirectory under Scalix mounted separately.

If you can have this server down for a bit, then run it manually,

rpm -Uvvh scalix-server*.rpm

This will show more detailed output about the error.

Thanks
Kev.

[jillrae]

OK Ran the install manually. Here is the output that showed up in the command line box:

linux:~/scalix-11.1.0-GA/software/scalix_server # rpm -Uvvh scalix-server*.rpm
D: ============== scalix-server-11.1.0.10849-1.sles10.i586.rpm
D: Expected size: 17080400 = lead(96)+sigs(180)+pad(4)+data(17080120)
D: Actual size: 17080400
D: scalix-server-11.1.0.10849-1.sles10.i586.rpm: Header SHA1 digest: OK (420ef9d 87f49c44f84a4820d8f88c181440d8596)
D: opening db environment /var/lib/rpm/Packages create:cdb:mpool:private
D: opening db index /var/lib/rpm/Packages rdonly mode=0x0
D: locked db index /var/lib/rpm/Packages
D: opening db index /var/lib/rpm/Name rdonly:nofsync mode=0x0
D: read h# 794 Header SHA1 digest: OK (31810863c9e0d88206c348ca89efb4c85d6b 5a64)
D: added binary package [0]
D: ============== scalix-server-11.1.0.10849-1.sles9.i586.rpm
D: Expected size: 17084935 = lead(96)+sigs(180)+pad(4)+data(17084655)
D: Actual size: 17084935
D: scalix-server-11.1.0.10849-1.sles9.i586.rpm: Header SHA1 digest: OK (e87084a3 016b355747a1ea6326efafd370a3c7f6)
warning: package scalix-server = 11.1.0.10849-1.sles10 was already added, skippi ng scalix-server < 11.1.0.10849-1.sles9
error: error reading from file scalix-server-11.1.0.10849-1.sles9.i586.rpm
D: closed db index /var/lib/rpm/Name
D: closed db index /var/lib/rpm/Packages
D: closed db environment /var/lib/rpm/Packages
D: May free Score board((nil))


I ran this also to check:

linux:~/scalix-11.1.0-GA/software/scalix_server # rpm -qa|grep scalix
scalix-platform-11.1.0.50-1
scalix-sis-11.1.0.45-1
scalix-server-11.0.3.10719-1.sles10
scalix-tomcat-connector-11.1.0.45-1.sles10
scalix-postgres-11.1.0.45-1
scalix-res-11.1.0.45-1
scalix-sac-11.1.0.45-1
scalix-mobile-11.1.0.45-1
scalix-tomcat-5.5.23-292
scalix-swa-11.1.0.51-1
scalix-text-extractors-1.0-1.sles10
linux:~/scalix-11.1.0-GA/software/scalix_server #

The server still didn't install. H-m-m-m-m

I checked the install log, but nothing new was written in there.

[jillrae]

OK. Revisited the manual install. I saw it tried to install both SLES 10 and SLES 9 version. Re-typed the rpm command, but this time only specified the SLES 10 package. It ran this time. Everything had YES or OK until this part:

D: ========== +++ scalix-server-11.1.0.10849-1.sles10 i586-linux 0x0
D: Expected size: 17080400 = lead(96)+sigs(180)+pad(4)+data(17080120)
D: Actual size: 17080400
D: scalix-server-11.1.0.10849-1.sles10: Header SHA1 digest: OK (420ef9d87f49c44f84a4820d8f88c181440d8596)
D: install: scalix-server-11.1.0.10849-1.sles10 has 6856 files, test = 0
D: install: %pre(scalix-server-11.1.0.10849-1.sles10.i586) synchronous scriptlet start
D: install: %pre(scalix-server-11.1.0.10849-1.sles10.i586) execv(/bin/sh) pid 32670
+ log=/tmp/ominstall.log
+ Olog=/tmp/ominstall.Olog
+ '[' -f /tmp/ominstall.log ']'
+ touch /tmp/ominstall.Olog
+ chmod 666 /tmp/ominstall.Olog
+ cat /tmp/ominstall.log
+ rm -f /tmp/ominstall.log
+ touch /tmp/ominstall.log
+ chmod 666 /tmp/ominstall.log
+ /usr/sbin/groupadd -r scalix
+ PATH=/opt/scalix/bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/X11R6/bin
+ export OM_UNI_DO_NOT_LOAD=TRUE
+ OM_UNI_DO_NOT_LOAD=TRUE
+ /etc/init.d/scalix stop
+ '[' -f /opt/scalix/bin/omsetsvc ']'
+ '[' -f /opt/scalix/global/config ']'
+ '[' -f /etc/opt/scalix/instance.cfg ']'
++ omsetsvc -p all
+ test -n ''
+ '[' '!' -f /etc/opt/scalix/instance.cfg ']'
+ '[' -x /opt/scalix/bin/omcheckgc ']'
++ wc -l
++ omcheckgc -l
+ num_instances=1
+ '[' 1 -eq 1 ']'
++ omrealpath '~/'
+ datadir=/var/opt/scalix/lx/s/
+ '[' /var/opt/scalix/lx/s/ '!=' /var/opt/scalix/ ']'
+ echo 'Cannot automatically update the Scalix' 'Server data directories, as they are in a non standard place'
+ exit 10
D: install: waitpid(32670) rc 32670 status a00 secs 0.378
error: %pre(scalix-server-11.1.0.10849-1.sles10.i586) scriptlet failed, exit status 10
error: install: %pre scriptlet failed (2), skipping scalix-server-11.1.0.10849-1.sles10
D: ========== --- scalix-server-11.0.3.10719-1.sles10 i586-linux 0x0
D: running post-transaction scripts
D: closed db index /var/lib/rpm/Basenames
D: closed db index /var/lib/rpm/Name
D: closed db index /var/lib/rpm/Packages
D: closed db environment /var/lib/rpm/Packages


Some kind of a scriptlet failure.

[KevinAnderson]

Can you run ompatchom first?

It looks like there are some incorrect entries about where you data is. That should fix them. If not, can you please verify that you have data in /var/opt/scalix/??/s, but not in /var/opt/scalix?

Then verify that the contents of /opt/scalix/global/config. We mostly just care about the datadir and the hostname.

Kev.

[jillrae]

Tried to run the ompatchom command, but got that it wasn't installed. so checked the directory structure.

Instead of /var/op/scalix/lx/s/ [lots of scalix folders] I have /var/opt/scalix/lx/s/lx/[lots of sclaix folders]

I'm pretty sure before the upgrade all the scalix folders were under the/var/opt/scalix/lx/s directory. Wonder where the extra /lx/ layer came from?

The /opt/scalix/global/config shows

OMNAME = linux
OMDATADIR = /var/opt/scalix/lx/s

So I moved all the scalix folders up one directory to the correct place and re-ran the manual isntall. A querry of the installed packages showed the server now at the 11.1 version. Will continue the check out the install to see if it runs Ok.

thanks

jillrae

[jillrae]

OK started checking out the server. Had lots of problems, so uninstalled and then reinstalled. The re-installtion went great until it tried to configure the web apps. The installer log shows:

2007-06-20 15:37:59,393 CRITICAL Operation failed: Web apps config function configure_res failed: Failed to change password for "query admin" user sxqueryadmin@linux.garrettcontainer.com. /opt/scalix/bin/ommodu reported the following error: ommodu : [OM 8040] The user can't be found, or has been deleted.
.
2007-06-20 15:37:59,394 ERROR Operation failed: Aborting web apps configuration

SO went to command line and tried to create the user sxqueryadmin. It said sxqueryadmin already existed. So I changed its password. Came back and said changed success ful.

Tired running the installer and re-config the install. Same error when configuring the web apps.

jillrae

[KevinAnderson]

Omdelu the queryadmin user.
Uninstall everything, leave the mailstore, but delete everything else.
delete -R /etc/opt/scalix*
delete -R /opt/scalix*
verify that scalix data is where it belongs (/var/opt/scalix/??/s)
reinstall.

Note, this mean regenerating the cache and reindexing all of your email. If you want to avoid that, you can try keeping the postgres data too.

Kev.

[jillrae]

Tried to dele the sxqueryadmin account. Came back and said the user doe not exit or has been deleted.

Did a complete uninstall, just left mail store. Deleted the 2 folders you listed also. Doulbe checked and data was under the s/ directory.

Did a new install. Failed at same point when tring to configure the webmail apps. Same problem, "failed to change pass word for sxqueryadmin...."

Just for fun I did omshowu -n sxqueryadmin.

linux:~ # omshowu -n sxqueryadmin
Authentication ID: sxqueryadmin@garrettcontainer.com
Globally Unique ID: 18800000b5943734-3.0.861.291
User Name : sxqueryadmin /CN=sxqueryadmin
MailNode : gcsmail,garrettcontaine
Internet Address : sxqueryadmin@garrettcontainer.com
System Login : 60534
Password : set
Admin Capabilities : NO
Mailbox Admin Capabilities : NO
Language : C
Mail Account: Unlocked
Last Signon : Never.
Receipt of mail : ENABLED
Service level : 0
Excluded from Tidying : NO
Recovery Folder visible : NO
User Class : Full
SIS URL : sxidx://linux.garrettcontainer.com/0880 ... .0.861.291

Why doesn't it have any admin capibilities? Should it?

jillrae

[KevinAnderson]

The logs you posted above show a different name than what you're seeing here.

Authentication ID: sxqueryadmin@garrettcontainer.com
vs.
Failed to change password for "query admin" user sxqueryadmin@linux.garrettcontainer.com.

Can you modify the authID to match what it's looking for?

Kev.

[jillrae]

Oh, I didn't notice that. Of course I'll change the authID. How do you do it?

Thanks

jillrae

[KevinAnderson]

ommodu should let you do it.

ommodu <authid> --authid <newauthid>

Kev.

[jillrae]

I changed the authID. It worked fine. did a re-config on the installtion. All configured OK. YEA!!!!

Logged in successfully using my personal email account. But could not log in using the sxadmin account. Says I don't have the correct user ID or password. I will work on this one.

Thank you.

jillrae