Re: [Sks-devel] DB_INIT_LOCK problems?

[Jason Harris]

On Fri, Jul 23, 2004 at 08:25:02PM -0400, Jason Harris wrote:
> On Fri, Jul 23, 2004 at 06:15:25PM -0400, Yaron Minsky wrote:

> > I was never able to get it to work quite properly.  When I enable it
> > with a recent version of SKS, everything locks up, though I don't
> > remember the details.  I suspect this is because I'm doing something
> > wrong with the way I deal with Berkeley DB.  But I've never been able
> > to suss it out.
> 
> I've enabled it and am running "db[42]_deadlock -t 1.0" from/on the
> KDB database directory/environment.  Starting db_deadlock before either
> sks process starts has kept my SKS processes running without problems/
> complaints for over 4 hours now.
> 
> Some Perl code that reads the key db while SKS is running has gotten
> DB_LOCK_DEADLOCK/EAGAIN/35 on some $cursor->c_get() calls, but trying
> the operation again after a small sleep() works fine.  (Note that I
> haven't checked the OCAML code/libraries for this behavior, however.)

Well, deadlock is hazardous to SKS - both "sks db" and "sks recon" 
were wedged by this (TZ=UTC):

  2004-07-24 21:31:51 Del'ng hash 94318B9CBA39316D1B1E95E83E1E1AA0
  2004-07-24 21:31:51 Adding hash F7CAD50EFFDC22AA0C7F6FDE721BFC61
  2004-07-24 21:33:22 Fatal database error: Bdb.DBError("caml_db_put: 
DB_LOCK_DEADLOCK: Locker killed to resolve a deadlock")
  2004-07-24 21:33:22 Key addition failed: Eventloop.SigAlarm
  2004-07-24 21:33:22 Handling /pks/hashquery

-- 
Jason Harris           |  NIC:  JH329, PGP:  This _is_ PGP-signed, isn't it?
address@hidden _|_ web:  http://keyserver.kjsl.com/~jharris/
          Got photons?   (TM), (C) 2004

pgpJY5f7Kpddu.pgp
Description: PGP signature