[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Rerunning rejected replog fails (by design?)

--On Thursday, November 04, 2004 11:31 AM -0800 Todd Lyons <tlyons@ivenue.com> wrote:

The subject is probably a little misleading.  It doesn't fail, the
replication is just skipped because the entries are too "old".  I found
someone who complained of this back in September of this year as well.
The solution as I found it was to:
1) backup all replog info to someplace safe
2) shut down slurpd daemon
3) delete slurpd.status and slurpd.replog
4) slurpd -r {slavename}*.rej -o (I used -d -1 to see everything)
5) start up slurpd daemon
6) check data exists in all locations where it should

Patches should be submitted via the ITS system, not to the software list.


As for the issue you saw, all you need to do is move the reject pieces to a different directory, and then use slurpd in oneshot mode, rather than doing the 6 steps you did...


Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html