Leap 42.1 citadel service will not start....................

I have tried Opensuse 13.2 and Leap 42.1 and both have resulted in the same result:

citadel.service - Citadel Groupware Server Loaded: loaded (/usr/lib/systemd/system/citadel.service; enabled) Active: failed (Result: exit-code) since Mon 2016-02-08 15:59:19 CST; 55s ago Process: 29576 ExecStart=/usr/sbin/citserver -d -l${CITADEL_LOG_FACILITY} ${CITADEL_DEBUG_AREA} ${CITADEL_HOME_DIR} (code=exited, status=0/SUCCESS) Main PID: 29577 (code=exited, status=105)
Feb 08 15:59:19 Server-Linux citserver[29578]: bdb(): Setting up DB environment Feb 08 15:59:19 Server-Linux citserver[29578]: dbenv->open(dbenv, /var/lib/citadel/data/, 10899, 0) Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Finding last valid log LSN: file: 1 offset 114165 Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Recovery starting from [1][112950] Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Recovery complete at Mon Feb 8 15:59:19 2016 Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Maximum transaction ID 80000019 Recovery checkpoint [1][114165] Feb 08 15:59:19 Server-Linux citserver[29578]: Starting up DB Feb 08 15:59:19 Server-Linux citserver[29578]: Checking floor reference counts Feb 08 15:59:19 Server-Linux citserver[29578]: bdb(): cursor still in progress on cdb 04: can’t begin transaction during r/o cursor Feb 08 15:59:19 Server-Linux citserver[29578]: citserver is stopping in order to prevent data loss. uid=0 gid=0 euid=0 egid=0

I have tried compiling citadel from source using the easy installer from citadel and adding the repo’s from OBS. Both fail at the same point.

Does anyone have any idea why this is failing on a fresh install?

citadel.service - Citadel Groupware Server Loaded: loaded (/usr/lib/systemd/system/citadel.service; enabled) Active: failed (Result: exit-code) since Mon 2016-02-08 15:59:19 CST; 55s ago Process: 29576 ExecStart=/usr/sbin/citserver -d -l${CITADEL_LOG_FACILITY} ${CITADEL_DEBUG_AREA} ${CITADEL_HOME_DIR} (code=exited, status=0/SUCCESS) Main PID: 29577 (code=exited, status=105) 
Feb 08 15:59:19 Server-Linux citserver[29578]: bdb(): Setting up DB environment Feb 08 15:59:19 Server-Linux citserver[29578]: dbenv->open(dbenv, /var/lib/citadel/data/, 10899, 0) Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Finding last valid log LSN: file: 1 offset 114165 Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Recovery starting from [1][112950] Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Recovery complete at Mon Feb 8 15:59:19 2016 Feb 08 15:59:19 Server-Linux citserver[29578]: DB: Maximum transaction ID 80000019 Recovery checkpoint [1][114165] Feb 08 15:59:19 Server-Linux citserver[29578]: Starting up DB Feb 08 15:59:19 Server-Linux citserver[29578]: Checking floor reference counts Feb 08 15:59:19 Server-Linux citserver[29578]: bdb(): cursor still in progress on cdb 04: can't begin transaction during r/o cursor Feb 08 15:59:19 Server-Linux citserver[29578]: citserver is stopping in order to prevent data loss. uid=0 gid=0 euid=0 egid=0

I guess no one else is having any services issues with 13.2 or Leap 42.1. This is with fresh install of both and using the repo’s that where setup. So neither the Citadel folks or the Opensuse community have any ideas? Back to the drawing board.

Hi
In your output I see r/o read-only? If so, check the permissions/ownership down in the relevant directories…

I know nothing about Citadel.

But, based on the log entries you posted, the database was found to be corrupted and a repair executed successfully (Keep in mind a successful database repair doesn’t always mean a functioning database).

So, the big question to start with is how important that old data is.
If it’s critical, then it’ll be worthwhile to troubleshoot.
If your business is tiny and the data not important (eg a relatively new install with almost no history), then it’ll probably make more sense to junk the data and do a brand new install.
In any case, good database practice generally does not recommend using a repaired database. Typical Best Practice is to install new and import the data from the repaired database into the new database created by the new Install.

In any case, at least what you’ve posted suggests that Citadel started up and is attempting to run but is failing due to a Citadel app problem, not an openSUSE problem (sorry Malcolm, I don’t see where there is a read-only permissions problem).

You <might> find more appropriate help in a Citadel Help forum, if you need help extracting log information or verifying proper install we can provide that here… But, to me it looks like your Citadel app is running… or at least trying to run properly but the database is likely still a problem.

TSU

Re-reading your original post saying this is a new install,

It’s fairly obvious then that there is a fundamental problem with the way the install was written.
Without actually trying to install, I’d most likely guess that the install is using old versions of apps, particualrly the database app.

Next time you post, include proper details so that people can state authoritatively what problems there are instead of speculating…

For example,
You should have included…

  • The OBS URL.
  • If you compile from source, provide the reference and download links.
  • If the install fails, post the entire install log or stdout. If too large to post in a Forum post, post on a pastebin and provide a link in your Forum post. ONLY if you’re very, very sure that you’re not leaving anything out, you can post only the error but often the Poster doesn’t have that expertise to know what can be excluded so when in doubt, post the entire thing.

Recommendation if you want to bypass troubleshooting…
Many OBS builds are available as virtualized appliances. If this is the case, then download the appliance and deploy on a virtualed platform.

TSU