Adding a Chunk in a Cluster
A user with two AIX servers in a failover cluster (SAN storage mirrored via AIX LVM) asked how to add a chunk to a dbspace: whether he needs to create a filesystem on each storage, and whether the chunks/dbspaces would be visible if the standby takes over. Replies noted that with disk/SAN-level mirroring this generally works, but that the usual setup is a single shared storage volume mounted only on the active node, so you add the dbspace/chunk once on the master and the new master sees it after failover. Art Kagel said he trusts Informix HDR more than disk-mirror failover clustering because of the out-of-sync window, and asked for clarification of the secondary's type (cold, HDR, RSS, SDS). The poster never clarified, so no definitive resolution is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
I have 2 servers in a fail over cluster and 1 have 2 storages. One storage makes mirror to the other one, And .... I need to add a chunk to a dbspace. I think I need to make 2 file systems One for each storage and add the chunk to the dbspace. I think it will work ok with the primary server but what will happend if the prymary server fail and the second server takes the control, will be available all the chunks and dbspaces to the second server? Thanks in advance. Enviado desde mi iPhone --Apple-Mail-3A4C10DF-FACC-4413-9991-A555B63BA707
A failover cluster as in only one is online at a time and the disks it will look at at failover time are mirrored from one system to the other using disk or SAN level mirroring? That should work. Honestly, I trust Informix HDR far more than any disk mirror based failover clustering technology. There is too big a window for the mirrored drives to be too far out-or-sync for the failover instance of Informix to be able to successfully recover. Art Art S. Kagel, Principal Consultant ASK Database Management Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Fri, Jun 20, 2014 at 2:54 PM, Jorge <jorgervt@gmail.com> wrote: > I have 2 servers in a fail over cluster and 1 have 2 storages. One storage > makes mirror to the other one, > And .... > I need to add a chunk to a dbspace. > I think I need to make 2 file systems > One for each storage and add the chunk to the dbspace. > I think it will work ok with the primary server but what will happend if > the > prymary server fail and the second server takes the control, will be > available > all the chunks and dbspaces to the second server? > > Thanks in advance. > > Enviado desde mi iPhone > --Apple-Mail-3A4C10DF-FACC-4413-9991-A555B63BA707 > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c2ae0e0be34f04fc498036
We have a SAN, and the mirror is trought de LVM of the AIX. Enviado desde mi iPhone El 20/06/2014, a las 12:28, "Art Kagel" <art.kagel@gmail.com> escribió: > A failover cluster as in only one is online at a time and the disks it will > look at at failover time are mirrored from one system to the other using > disk or SAN level mirroring? > > That should work. Honestly, I trust Informix HDR far more than any disk > mirror based failover clustering technology. There is too big a window for > the mirrored drives to be too far out-or-sync for the failover instance of > Informix to be able to successfully recover. > > Art > > Art S. Kagel, Principal Consultant > ASK Database Management > > Blog: http://informix-myview.blogspot.com/ > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and do not reflect on the IIUG, nor any other organization with which I am > associated either explicitly, implicitly, or by inference. Neither do > those opinions reflect those of other individuals affiliated with any > entity with which I am affiliated nor those of the entities themselves. > > On Fri, Jun 20, 2014 at 2:54 PM, Jorge <jorgervt@gmail.com> wrote: > >> I have 2 servers in a fail over cluster and 1 have 2 storages. One storage >> makes mirror to the other one, >> And .... >> I need to add a chunk to a dbspace. >> I think I need to make 2 file systems >> One for each storage and add the chunk to the dbspace. >> I think it will work ok with the primary server but what will happend if >> the >> prymary server fail and the second server takes the control, will be >> available >> all the chunks and dbspaces to the second server? >> >> Thanks in advance. >> >> Enviado desde mi iPhone >> --Apple-Mail-3A4C10DF-FACC-4413-9991-A555B63BA707 > ******************************************************************************* >> Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c2ae0e0be34f04fc498036 > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >
I thing you have to check well how your configuration is in that particular Cluster. The usual situation should be to have one shared storage volume to use with database spaces. This is one storage which is only active on the masater node of cluster. This implies have only to create the dbspace in the masater, When cluster do a failover o sweitchover, te new master node will see mounted the shared volume. Greetings. --Nestor On vie, 2014-06-20 at 14:54 -0400, Jorge wrote: > I have 2 servers in a fail over cluster and 1 have 2 storages. One storage > makes mirror to the other one, > And .... > I need to add a chunk to a dbspace. > I think I need to make 2 file systems > One for each storage and add the chunk to the dbspace. > I think it will work ok with the primary server but what will happend if the > prymary server fail and the second server takes the control, will be available > all the chunks and dbspaces to the second server? > > Thanks in advance. > > Enviado desde mi iPhone > --Apple-Mail-3A4C10DF-FACC-4413-9991-A555B63BA707 > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >
OK. Is the secondary a cold secondary accessing the mirror? Is it an HDR or RSS secondary? An SDS secondary? Sorry I don't understand your configuration so I can't answer the question. Art Art S. Kagel, Principal Consultant ASK Database Management Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Fri, Jun 20, 2014 at 5:23 PM, Jorge <jorgervt@gmail.com> wrote: > We have a SAN, and the mirror is trought de LVM of the AIX. > > Enviado desde mi iPhone > > El 20/06/2014, a las 12:28, "Art Kagel" <art.kagel@gmail.com> escribió: > > > A failover cluster as in only one is online at a time and the disks it > will > > look at at failover time are mirrored from one system to the other using > > disk or SAN level mirroring? > > > > That should work. Honestly, I trust Informix HDR far more than any disk > > mirror based failover clustering technology. There is too big a window > for > > the mirrored drives to be too far out-or-sync for the failover instance > of > > Informix to be able to successfully recover. > > > > Art > > > > Art S. Kagel, Principal Consultant > > ASK Database Management > > > > Blog: http://informix-myview.blogspot.com/ > > > > Disclaimer: Please keep in mind that my own opinions are my own opinions > > and do not reflect on the IIUG, nor any other organization with which I > am > > associated either explicitly, implicitly, or by inference. Neither do > > those opinions reflect those of other individuals affiliated with any > > entity with which I am affiliated nor those of the entities themselves. > > > > On Fri, Jun 20, 2014 at 2:54 PM, Jorge <jorgervt@gmail.com> wrote: > > > >> I have 2 servers in a fail over cluster and 1 have 2 storages. One > storage > >> makes mirror to the other one, > >> And .... > >> I need to add a chunk to a dbspace. > >> I think I need to make 2 file systems > >> One for each storage and add the chunk to the dbspace. > >> I think it will work ok with the primary server but what will happend if > >> the > >> prymary server fail and the second server takes the control, will be > >> available > >> all the chunks and dbspaces to the second server? > >> > >> Thanks in advance. > >> > >> Enviado desde mi iPhone > >> --Apple-Mail-3A4C10DF-FACC-4413-9991-A555B63BA707 > > > > ******************************************************************************* > >> Forum Note: Use "Reply" to post a response in the discussion forum. > > > > --001a11c2ae0e0be34f04fc498036 > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --089e0160bbeec3107704fc4bb991