Hi,
We run RT 3.0.10 today and now we want to upgrade to the newest version.
Does anybody know if it’s goung to be a problem to just run an upgrade?
Even if the new version goes from 3.0 to 3.4?
I am really scared that RT or the database will crash so I lose data.
Hilde
We run RT 3.0.10 today and now we want to upgrade to the newest version.
Does anybody know if it’s goung to be a problem to just run an upgrade?
Even if the new version goes from 3.0 to 3.4?
I upgraded from 3.0.?? to 3.4.1 on Debian and it went smoothly.
I am really scared that RT or the database will crash so I lose data.
Don’t be. I recently upgraded and had no problems whatsoever. But, just in case,
backup your database. Than there is no chance of losing any data at all.
– Pieter van der Spek, pieter@west.nl, West Consulting B.V., +31 15 2191 600
Hi
I’m on my way upgrading from 3.0.9 to 3.4.2 here…
and the RHEL4 glibc reports some bad things after upgrading:
[Tue Jul 05 02:21:56 2005] [warn] FastCGI: server
“/usr/local/rt3/bin/mason_handler.fcgi” restarted (pid 18153)
[Tue Jul 05 02:22:01 2005] [warn] FastCGI: server
“/usr/local/rt3/bin/mason_handler.fcgi” restarted (pid 18157)
*** glibc detected *** double free or corruption (!prev): 0x0b4c1180 ***
[Tue Jul 05 02:22:10 2005] [error] [client 192.168.0.17] FastCGI: incomplete
headers (0 bytes) received from server “/usr/local/rt3
/bin/mason_handler.fcgi”
[Tue Jul 05 02:22:10 2005] [warn] FastCGI: server
“/usr/local/rt3/bin/mason_handler.fcgi” (pid 17986) terminated due to uncaught
signal ‘6’ (Aborted)
or after I decided to try from scratch during install (make initialize-database)
we go here:
Creating ScripConditions…1.2.3.4.5.6.7.8.9.10.done.
Creating templates…1.*** glibc detected *** free(): invalid next size
(normal): 0x0ac49e18 ***
make: *** [initialize-database] Aborted
well, I’m stuck here - I tried to run at least initdb with “MALLOC_CHECK_=0”
which helps, but does not help really because instead of complaining it simply
segfaults when creating the templates…
I as a last resort moved to mysql 4.1.12 from mysql.com for rhel4 - well didn’t
help either here…
anyone having a working install here on rhel4?
or is it just some (new?) nasty perl-module that is used that breaks things
here?
any help appreciated - I’m pretty helpless at the moment…
thank you in advance
Harald Kapper, icq# 36178328 http://kapper.net
managing director, owner, loeblichgasse 6
chief executive officer 1090 vienna, .at
tel +43 1 3195500-0, fax +43 1 3195502, hk@kapper.net
--------------mark---------------
The following Statement is False.
The previous Statement is True.
- Welcome to my World -
Harald,
I’m having the same problem. Have you (or has anyone else) been able to find
a solution?I’m running CentOS 4 (essentially RHEL4) and have upgraded from RT 3.0.2 to
3.4.2, with FastCGI 2.4.2 and Apache 2.0.52-12. I have not done any Apache-,
MySQL-, or glibc-related updates since June 18th. RT and mod_fastcgi have to
be updated by hand. This problem started over the weekend on a system which
had been running stably for a few weeks, which leads me to suspect that a
logrotate-related process is somehow involved; however, I haven’t spotted
anything to confirm that yet.
James,
you should see segfaults of mason-handler in your httpd_error_log (imho).
but as Jesse already confirmed - upgrade your DBD::Mysql via CPAN to the latest
v3.001 and segfaults should go away again.
unlucky me migrated already everything in a 24h sweep to a new box (this time
debian). 
regards
Harald Kapper, icq# 36178328 http://kapper.net
software-development and loeblichgasse 6
security-consulting 1090 vienna, .at
tel +43 1 3195500-0, fax +43 1 3195502, hk@kapper.net
--------------mark---------------
The following Statement is False.
The previous Statement is True.
- Welcome to my World -