Re: External backup

On 27 July, 20:48, "Neil Truby" <neil.tr...@ardenta.com> wrote: > -----Original Message----- > From: Madison Pruet [mailto:mpru...@verizon.net] > Sent: 27 July 2009 19:38 > To: Neil Truby > Subject: Re: External backup > > Neil Truby wrote: > > IDS 10.0 for example, any O/S. > > > I've never understood why we have to an onmode -c block/onmode -c unblock > > when taking a clone of Informix's data disks for back-up purposes.  To my > > simple mind if Informix can fast recover from the original disks it should > > by definition be able to recover from the clone. > > onmode -c block flushes all dirty pages to disk and prevents any updates> > from happening.  This makes it possible to get a consistent copy of the > chunks. > > >> ***** In order to do fast recovery, it is a requirement to get a > > consistent copy of the disks. **** > > Thanks.  I follow all of this.  But if IDS can't reliably fast recover from > a set of cloned disks for these reasons, how can it fast recover reliably > from the *original* disks which by definition are identical to the clones > ...? If you are just copying then they may not be the same as writes are still happening to the first part of the disk after you copy it, since you copy the logs later the current position in the log will have moved hence those change will not be applied from the log. If you are splitting mirrors then the same thing applies, you split the log at a different time to the data hence they are out of sync and cannot be used to get to a consistent point in time. By splitting different devices at different points in time you effectively change the order of writes and hence break the consistency model. If you split the whole instance (logs+data) at the same time with no writes being applied anywhere whilst the split occurs then you should be ok in theory, it is safer though to just checkpoint and block for a few seconds.