Hi Francois,
as far as I know there is a limit on the number of conditions you can use (See elog:67303).
I guess you did hit that limit.

I think I've mentioned recently that ELOG is not a relational database, didn't I?
It is of course possible to drive screws with a hammer, but there do exist more suitable tools for that ;-)

Kind regards
Andreas

 
> Seriously, I really hope It could make it... Could you try it on your side ?

My personal opinion here is, that if you want others to investigate your problems, then the best way to do it is like that:
- attach a minimal configuration that reproduces your problem (never attach a 100 line configuration, unless you've tested that the problem disappears
regardless of which line you remove!);
- attach the entry data, if the behaviour depends on the data;
- use a specific, to the point subject line;
- explain what you did, what happened and what you would have expected to happen;
- ask kindly; and then
- wait and hope for the best ;-)

(This is actually a very general procedure; I think it is applicable to all newsgroups, forums, etc.)

BTW: This tip was absolutely free of charge ;-)

> I came accross the admin guide and I was reading / searching for a way to sync logbooks across sites...
> elogd mention "-m" and "-M" ... not elog but elogd... with that description :
> synchronize logbook(s) with remote server

If you would have followed the shown link to the "elogd.cfg syntax page", you would have found the chapter Mirroring:
https://midas.psi.ch/elog/config.html#mirroring

> Does it sync all logbooks ? 

Not necessarily, but that is the default.

> is there any examples somewhere or advice ?

See above: you'll find examples under #mirroring

> Thanks :)

You're welcome.

> elogd 3.1.0 moves all elog entries into year-named subdirectories. this feature makes it incompatible with older elogs and so should be clearly mentioned in the documentation,
> in the release announcement and in the release and migration notes. K.O.

That feature is one of the main reasons why the version jumped from 2.x to 3.x. 
A free tip: changes in major revisions do indicate some kind of incompatibility.
But yes, the release documentation by bitbucket is not really that useful: 
it is difficult for me too, to find out what changed with new releases. 
I have to admit here, that I haven't read any GIT tutorial yet.
By the way: you are welcome to contribute to the release documentation!

On your actual problem: to go back to a former version of ELOG you can simply
- stop elogd 3.X, 
- move all entries from the sub-directories one level up, and 
- start the 2.X version of elogd.

I wouldn't really call this an "incompatibility", would you? 
At least you can easily go back without much trouble.

Cheers
Andreas

> Stefan told me that the change was because some users were having thousands of yymmdda.log files
> in the logbook directories, and that sorting them into subdirectories by year at least did something to bring some 
> order.  Possibly to get around the lazy archivers, I suspect.

I'm actually the culprit, who did ask for it.

If you want to know the full story, here it is:
We have our logbook data of our accelerator operation logbooks on AFS (Andrew File System). 
And apparently AFS has a bloody stupid, hard coded limit: 
the total length of all file names in one directory cannot exceed 64k.
Our operation logbooks go back for more than a decade and do contain many, many, many attachment files.
One day - very unexpectedly - we did hit that limit. 
Removing temporary files (generated picture thumbnails) bought us time, and Stefan was nice enough to upgrade ELOG swiftly for us: a big "Thank You" to Stefan!