Re: Onbar Scalability Issues

Posted in 1999

Doug McAllister — Tue, 20 Jul 1999 14:12:20 -0400 — source: Informix-list mailing list archive (1991-1998)

Rick Rabe wrote: > This message is in MIME format. Since your mail reader does not understand > this format, some or all of this message may not be legible. > > ------_=_NextPart_001_01BECFD0.A496FABC > Content-Type: text/plain > > I'm not familiar with Netbackup, but use Networker on Siemens-Pyramid. My > BAR_MAX_BACKUP is set to 8. One thing I have found helpful is altering > bar_action, bar_instance, bar_object in the sysutils database to row-level > locking instead of page-level locking. > > HTH > rrabe > > > ---------- > > From: Steve Hawkins[SMTP:steve.hawkins@orange.co.uk] > > Reply To: Steve Hawkins > > Sent: Friday, July 09, 1999 3:38 AM > > To: informix-list@iiug.org > > Subject: Onbar Scalability Issues > > > > We are currently running informix 7.24 on Solaris 2.5.1 (E10000 > > Platform) and have been using onbar to do backups through to Netbackup. > > This generally works ok, despite the database being 300Gb of used pages > > in size, although it does fail with connection timeout errors. The > > connection timeout errors wouldn't be so bad if the connection to > > Netbackup was remote but in this case it is all on the same machine > > along with the backup robot. > > > > Also when trying to enable additional onbar's to run by specifiying the > > dbspaces to backup the error 54's get progressively worse to the point > > where a good daily backup is un-achievable. > > > > Does anyone else use onbar in this way i.e. onbar -b -L 0 <dbspace > > list>, and changing the onconfig to allow multiple onbar's to run > > (BAR_MAX_BACKUP 6) ? > > > I do it almost exactly this way. I have BAR_MAX_BACKUP set to 12 and the > same parm in Networker set to 12. I do an onbar -b -L 0 (with no dbspaces > specified). After the rootdbs is archived, onbar spawns 12 processes and > keeps 12 going in parallel until all of 53 of my dbspaces are complete. > Informix 7.30 on Solaris 2.6 and Networker 5.1 to two DLT drives. I archive > approx 50 gig in 45 minutes. > Doug... > > ------_=_NextPart_001_01BECFD0.A496FABC > Content-Type: text/html > Content-Transfer-Encoding: quoted-printable > > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> > <HTML> > <HEAD> > <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = > charset=3Dus-ascii"> > <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = > 5.5.2448.0"> > <TITLE>RE: Onbar Scalability Issues</TITLE> > </HEAD> > <BODY> > > I'm not = > familiar with Netbackup, but use Networker on Siemens-Pyramid. My = > BAR_MAX_BACKUP is set to 8. One thing I have found helpful is altering = > bar_action, bar_instance, bar_object in the sysutils database to = > row-level locking instead of page-level locking. > > HTH > rrabe > > <UL> > ---------- > From:   = > Steve = > Hawkins[SMTP:steve.hawkins@orange.co.uk] > Reply To: = >       Serif">Steve Hawkins > Sent:   = > Friday, July 09, 1999 3:38 = > AM > To: = >     Serif">informix-list@iiug.org > Subject: = >        Serif">Onbar Scalability Issues > > > We are currently running informix 7.24 = > on Solaris 2.5.1 (E10000 > Platform) and have been using onbar = > to do backups through to Netbackup. > This generally works ok, despite the = > database being 300Gb of used pages > in size, although it does fail with = > connection timeout errors. The > connection timeout errors wouldn't be = > so bad if the connection to > Netbackup was remote but in this case = > it is all on the same machine > along with the backup robot. > > > Also when trying to enable additional = > onbar's to run by specifiying the > dbspaces to backup the error 54's get = > progressively worse to the point > where a good daily backup is = > un-achievable. > > > Does anyone else use onbar in this way = > i.e. onbar -b -L 0 <dbspace > list>, and changing the onconfig = > to allow multiple onbar's to run > (BAR_MAX_BACKUP 6)  ? > > </UL> > </BODY> > </HTML> > ------_=_NextPart_001_01BECFD0.A496FABC--