We have been successfully using the TCK-##### for several months and suddenly today it has stopped incrementing. We were at ticket # 40 and this morning any new tickets come up as 12 and never increment? We have tried restarting the elog service. Any suggestions would be appreciated.

Thank You

Don

 A couple of comments:

• In principle you could use one elogd.cfg for two running daemons, but only if they do not modify the file. If you use "extensible attributes" for example, you would have write access. If this happens from both servers at the same time, the config file can get corrupt. This is even more true with a password file, which gets written every time a users accesses a page. If you only run one server at a time, then they can share the config file of course.
• The synchronization works bi-directional between two elog daemons. Requirement is that they have access to each other. The synchronization master acts like a "browser" to the slave to send/retrieve entries. So if you can affort punching a hole into your firewall between NET1 and NET2, you could synchronize two servers directly and don't have to bother with three servers. If you have however really physically separated nets, such that you need a memory stick to transport data from one net to the other, then indeed you need the above described scenario. I never tried it in this way, but in principle it should work.

Hi there. Just started recently set up the application and think its very usable and easily configured - great stuff

I'm looking to roll it out to the masses but have a slight logistical problem in that we work on 2 separate networks - NET1 with connection to the internet, and NET2 physically isolated from NET1 and the internet.

It would be desirable to allow users to add/update/delete logs from both NET1 and NET2 with periodic synchronisation.

I believe the mirroring facility is what should be used, and think the following setup and synchronisation steps may be workable:

Setup:

• elog installed on node of NET1
• elog installed on 2 nodes of NET2
• elog server running on NET1 and node1 of NET2 both using a common elogd.cfg

Performing Synchronisation:

1. Stop the server on NET1
2. copy logbook directories from NET1 to node2 of NET2
3. start server on node2 of NET2 using the -M option
4. once above complete, copy logbook directories from node 2 NET2 to NET1
5. Start the server on NET1

Am I on the right track?

Would very much appreciate your comments and advice on the above.

Cheers Dougie

Jul  9 19:05:00 elogd[27009]: Falling back to default group "elog"
Jul  9 19:05:00 elogd[27009]: Falling back to default user "elog"
Jul  9 19:05:02 elogd[27009]: Cannot restore original GID/UID.
Jul  9 19:05:02 elogd[27009]: Cannot remove pidfile "/var/run/elogd.pid" ; Permission denied
Jul  9 19:05:05 elogd[27013]: Falling back to default group "elog"
Jul  9 19:05:05 elogd[27013]: Falling back to default user "elog"
Jul  9 19:05:05 elogd[27013]: Cannot restore original GID/UID.
Jul  9 19:05:05 elogd[27013]: Cannot remove pidfile "/var/run/elogd.pid" ; Permission denied
Jul  9 19:05:05 elogd[27016]: Falling back to default group "elog"
Jul  9 19:05:05 elogd[27016]: Falling back to default user "elog"
Jul  9 19:05:06 elogd[27016]: Cannot restore original GID/UID.
Jul  9 19:05:06 elogd[27016]: Cannot remove pidfile "/var/run/elogd.pid" ; Permission denied

Jul  9 19:47:08 elogd[27335]: elogd server aborted.

Jul  9 19:47:16 elogd[27359]: elogd 2.7.3 built Apr 18 2008, 14:08:31 
Jul  9 19:47:16 elogd[27359]: revision 2104
Jul  9 19:47:16 elogd[27359]: Falling back to default group "elog"
Jul  9 19:47:16 elogd[27359]: Falling back to default user "elog"
Jul  9 19:47:16 elogd[27360]: Falling back to default group "elog"
Jul  9 19:47:16 elogd[27360]: Falling back to default user "elog"
Jul  9 19:47:16 elogd[27360]: Cannot restore original GID/UID.
Jul  9 19:47:16 elogd[27360]: Cannot remove pidfile "/var/run/elogd.pid" ; Permission denied
Jul  9 19:47:16 elogd[27359]: ImageMagick detected

Thanks a lot

Ok, I added sorting for "museremail" and "muserlist" type of attributes. Sorting is simply done by sorting the email addresses and first part of full name, respectively. The change is implemented in SVN revision #2115.

Ok, now I got the point. I fixed this problem in elog revision #2114. Quick filters are now working with icons: