[Csync2] Database backend is exceedingly busy => Terminating (requesting retry).

Art -kwaak- van Breemen ard+csync2 at telegraafnet.nl
Thu Mar 12 16:26:14 CET 2009


On Thu, Mar 12, 2009 at 11:22:11AM +0000, Keith Edmunds wrote:
> I'm getting the above message. Two questions:
> - does the "requesting retry" mean that this action will be retried in
> future? How many times?

Actually it means that you should retry the action. You will get
this problem if more than 1 csync2 process is accessing the
sqlite database, and there is some lock contention.
Processing hints with -c or updating neighbours with -u are
actions that can be repeated over and over until it is resolved.
The -h option is probably the only option that really needs to be
run again exactly as given.

> - what (if anything) can I do to work around this problem?

There is no real work around. For the incoming inetd section I
have scripted around it so that only the first connection runs
csync2 -i, for connectioncount > 1 it just exits, and the other
side will retry.

The only real solution is to rewrite csync in being a daemon that
has a backingstore (to do lock prevention), or to replace the SQL
part with a library that can deal with multiple access (like
using mysql for instance :-) ).

-- 
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in email?


More information about the Csync2 mailing list