[Linux-ha-dev] latest cts runs

2007-06-01 Thread Dejan Muhamedagic
Hello, Last nights Alan's odd collection of machines, which, for some reason, he calls lab cluster, ran CTS. It was rather successful. My very symmetrical 3-node xen cluster also showed similar results. There were two problems encountered: - failed resource audits, always during the NearQuorum t

Re: [Linux-ha-dev] latest cts runs

2007-06-01 Thread Lars Marowsky-Bree
On 2007-06-01T20:17:10, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > It looks like we could release soon. > > Anybody else wants to run a CTS? Now is the time if you have a > cluster to fiddle with. Which options do you think would be most helpful? How are you invoking CTS? It might be useful

Re: [Linux-ha-dev] latest cts runs

2007-06-04 Thread Dejan Muhamedagic
On Fri, Jun 01, 2007 at 08:23:47PM +0200, Lars Marowsky-Bree wrote: > On 2007-06-01T20:17:10, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > > > It looks like we could release soon. > > > > Anybody else wants to run a CTS? Now is the time if you have a > > cluster to fiddle with. > > Which optio

Re: [Linux-ha-dev] latest cts runs

2007-06-04 Thread Lars Marowsky-Bree
On 2007-06-04T15:25:27, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > No special options: > > python /usr/lib/heartbeat/cts/CTSlab.py -v2 --logfile /var/log/ha-log > --stonith no 500 > ctsrun.out 2>&1 > > I use my own configuration, which consists of six groups and each > group with two IP add

Re: [Linux-ha-dev] latest cts runs

2007-06-04 Thread Dejan Muhamedagic
On Mon, Jun 04, 2007 at 03:48:21PM +0200, Lars Marowsky-Bree wrote: > On 2007-06-04T15:25:27, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > > > No special options: > > > > python /usr/lib/heartbeat/cts/CTSlab.py -v2 --logfile /var/log/ha-log > > --stonith no 500 > ctsrun.out 2>&1 > > > > I use

Re: [Linux-ha-dev] latest cts runs

2007-06-04 Thread Andrew Beekhof
On 6/4/07, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: On Mon, Jun 04, 2007 at 03:48:21PM +0200, Lars Marowsky-Bree wrote: > On 2007-06-04T15:25:27, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > > > No special options: > > > > python /usr/lib/heartbeat/cts/CTSlab.py -v2 --logfile /var/log/ha-lo

Re: [Linux-ha-dev] latest cts runs

2007-06-04 Thread Lars Marowsky-Bree
On 2007-06-04T16:21:56, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > Aha. Interestingly, it does on the Alan's cluster: > > FLAGS="--logfile /var/log/ha-log-$F -2 --trunc --clobber-cib -r --facility $F > --stonith 1 --fencing 0" Oh. I was missing the --clobber-cib, I think. I'll start up the

Re: [Linux-ha-dev] latest cts runs

2007-06-06 Thread Lars Marowsky-Bree
On 2007-06-01T20:17:10, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > Last nights Alan's odd collection of machines, which, for some > reason, he calls lab cluster, ran CTS. It was rather successful. > My very symmetrical 3-node xen cluster also showed similar > results. With 3 nodes, my cluster

Re: [Linux-ha-dev] latest cts runs

2007-06-07 Thread Andrew Beekhof
On 6/6/07, Lars Marowsky-Bree <[EMAIL PROTECTED]> wrote: On 2007-06-01T20:17:10, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > Last nights Alan's odd collection of machines, which, for some > reason, he calls lab cluster, ran CTS. It was rather successful. > My very symmetrical 3-node xen clust

Re: [Linux-ha-dev] latest cts runs

2007-06-07 Thread Lars Marowsky-Bree
On 2007-06-07T11:05:16, Andrew Beekhof <[EMAIL PROTECTED]> wrote: > >(I'm looking forward to the cleanup which the openAIS integration will > >bring here, as making that module pluggable should help untangle it ;-) > the openais version wont be using HA_Message* to send cluster messages > so i don

Re: [Linux-ha-dev] latest cts runs

2007-06-07 Thread Dejan Muhamedagic
On Wed, Jun 06, 2007 at 11:38:56PM +0200, Lars Marowsky-Bree wrote: > On 2007-06-01T20:17:10, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > > > Last nights Alan's odd collection of machines, which, for some > > reason, he calls lab cluster, ran CTS. It was rather successful. > > My very symmetric

Re: [Linux-ha-dev] latest cts runs

2007-06-07 Thread Lars Marowsky-Bree
On 2007-06-07T15:16:18, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > Can you send over the log for that one. I'm collecting those ;-) What are you using the logs for? I can try to find them, I hope I didn't delete them yet. > > I _think_ the second cheapest path would be to move the compressi

Re: [Linux-ha-dev] latest cts runs

2007-06-08 Thread Dejan Muhamedagic
On Fri, Jun 08, 2007 at 01:00:14AM +0200, Lars Marowsky-Bree wrote: > On 2007-06-07T15:16:18, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > > > Can you send over the log for that one. I'm collecting those ;-) > > What are you using the logs for? Hang them in my living room ;-) Sometimes also tr

Re: [Linux-ha-dev] latest cts runs

2007-06-08 Thread Lars Marowsky-Bree
On 2007-06-08T11:06:02, Dejan Muhamedagic <[EMAIL PROTECTED]> wrote: > > Right. That's why the compression needs to be done in the *_add call > > directly, so that the lower levels need not know about message > > compression. > > > > Hinting is broken by design, and I don't want to see that intro