Hi Dejan, I confirmed it in the environment where NLS_LANG was set in Japanese(Japanese_Japan.AL32UTF8).
I changed the expiration date of the OCFMON user and pushed forward the date of the system for one year. I confirmed that the next processing worked definitely.(...on oracle12c) Confirmed 1) After OCFMON user became expired (EXPIRED), the monitor processing in the sysdba user succeeds. Confirmed 2) The grep judgment of the EXPIRED character string is carried out definitely. Confirmed 3) When we start oracle again after OCFMON user expired, the time limit of the OCFMON user is changed. > > 415 if echo "$output" | grep -w "EXPIRED" >> /dev/null; then > > Also, could you verify if common_sql_filter() need modifications? As a result, the correction of the next grep was not necessary.(Confirmed 2,Confirmed 3) Best Regards, Hideo Yamauchi. ----- Original Message ----- > From: "[email protected]" <[email protected]> > To: Dejan Muhamedagic <[email protected]>; High-Availability Linux > Development List <[email protected]> > Cc: > Date: 2014/7/22, Tue 20:50 > Subject: Re: [Linux-ha-dev] [Question] About the change of the oracle > resource agent. > > Hi Dejan, > > All right!! > >> Is that with the latest version? > > > I confirm RA now in Oracle12c. > It is the latest edition of oracle. > > Many Thanks! > Hideo Yamauchi. > > > > ----- Original Message ----- >> From: Dejan Muhamedagic <[email protected]> >> To: [email protected]; High-Availability Linux Development List > <[email protected]> >> Cc: >> Date: 2014/7/22, Tue 18:46 >> Subject: Re: [Linux-ha-dev] [Question] About the change of the oracle > resource agent. >> >> Hi Hideo-san, >> >> On Tue, Jul 22, 2014 at 11:07:29AM +0900, [email protected] wrote: >>> Hi All, >>> >>> I am going to explain the next change to our user. >>> >>> * https://github.com/ClusterLabs/resource-agents/pull/367 >>> * https://github.com/ClusterLabs/resource-agents/pull/439 >>> >>> >>> Let me confirm whether it is the next contents that a patch intends. >>> >>> 1) Because it was a problem that OCFMON user was added while the > oracle >> manager did not know it, patch changed it to appoint it explicitly. >> >> The OCFMON user and password parameters are optional, hence in >> this respect nothing really changed. The user is still created >> by the RA. However, it is good that they're now visible in the >> meta-data. >> >>> 2) Patch changed a deadline of OCFMON.(A deadline for password of the >> default may be 180 days.) >> >> That's the problem we had with the previous version. Now there's >> a profile created for the monitoring user which has unlimited >> password expiry. If the password expired in the meantime, due to >> a missing profile, then it is reset. >> >> If the monitor still fails, the RA tries as sysdba again. >> >>> 3) Patch kept compatibility with old RA. >> >> Yes. >> >>> Is there the main point of any other patches? >> >> No. >> >>> If there is really the problem that occurred, before this change, > please >> teach to me. >> >> As mentioned above, the issue was that the password could >> expire. >> >>> I intend to really show the problem that happened to a user. >>> * For example, a time limit of OCFMON expired and failed in a monitor > of >> oracle >> >> Is that with the latest version? >> >> Cheers, >> >> Dejan >> >>> I am going to send a patch later. >>> >>> Best Regards, >>> Hideo Yamauchi. >>> _______________________________________________________ >>> Linux-HA-Dev: [email protected] >>> http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev >>> Home Page: http://linux-ha.org/ >> > _______________________________________________________ > Linux-HA-Dev: [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev > Home Page: http://linux-ha.org/ > _______________________________________________________ Linux-HA-Dev: [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev Home Page: http://linux-ha.org/
