HIDEO - Please give a try to the attached patch and let me know what you think. I put back the old mechanism of checking process status without using pgrep and also cleaned a couple of other things.
Thanks. Serge. On Feb 15, 2008 4:28 PM, Serge Dubrouski <[EMAIL PROTECTED]> wrote: > On Thu, Feb 14, 2008 at 7:36 PM, HIDEO YAMAUCHI > <[EMAIL PROTECTED]> wrote: > > > Hi Serge, > > > > We discussed two problems. > > > > 1)Can the process done swap out of do detection in pgrep? > > There is a doubtful mention in online help of pgrep. > > 2)There is the case that there is a process having same PID in a failover > > node. > > For example, this problem get up accidentally when an active node > > rebooted suddenly. > > > > In conclusion... > > > > 1)Can detect the process done swap out of in pgrep. > > But when process became the zombie process, pgrep cannot detect it, but > > we decided not to consider > > this. > > 2)When there was a PID file, we decided to perform the check of PID for a > > result of pgrep. > > And, the enforcement of kill decided to abolish it because there were not > > many meanings. > > > > I revised the mistake of some patches which you offered. > > > > There may be waste to this script. > > Remove waste and I wish you include it in a development version. > > > > Sorry... > > This revision does not consider the contents of the other mailing lists > > that you discuss. > > (http://www.gossamer-threads.com/lists/linuxha/users/45882) > > > > Hi HIDEO - > > Thanks for your help. It turned out that I kind of misunderstood OCF > requirements and in fact there is no real need to use pgrep and I can > fall back to "kill -0" schema to check if a process is up. I'll be > busy with some other stuff for a couple of days then I'll provide > little bit different patch that shouldn't have all those pgrep > problems. > > Thanks again. > > -- > Serge Dubrouski. > -- Serge Dubrouski.
pgsql.patch
Description: Binary data
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
