Archive Behavior Change???

Good Morning, IDS 11.50.FC9X1 OS AIS 6.1 Recently I implemented a tiered approach to my onbar Informix Archives. IE... I started doing level 1's in bwtween the level 0's. I have multiple systems here, most of which are still on 11.50.FC5 (all AIX) however on my only 11.50.FC9X1 system I get the following message on the 2nd consecutive day of level 1 archive: 2011-11-22 18:00:14.333852 549006 463030 Unable to start the storage space backup: The existing level-0 backup for DBspace r1 is too old to allow any incremental backup.. I use the exact same schedule on my 11.50.FC5 systems with no implications. Does anyone know if this behavior has changed between the two versions? Oh yeah, the same storage manager is used for all systems. It is ADSM 5.3.2. Thanx, Dan

original post: Good Morning, IDS 11.50.FC9X1 OS AIS 6.1 Recently I implemented a tiered approach to my onbar Informix Archives. IE... I started doing level 1's in bwtween the level 0's. I have multiple systems here, most of which are still on 11.50.FC5 (all AIX) however on my only 11.50.FC9X1 system I get the following message on the 2nd consecutive day of level 1 archive: 2011-11-22 18:00:14.333852 549006 463030 Unable to start the storage space backup: The existing level-0 backup for DBspace r1 is too old to allow any incremental backup.. I use the exact same schedule on my 11.50.FC5 systems with no implications. Does anyone know if this behavior has changed between the two versions? Oh yeah, the same storage manager is used for all systems. It is ADSM 5.3.2. Thanx, Dan Response: Well that message comes from a time stamp comparison of the time stamp of the last level 0 of the dbspace, compared to the current time stamp on the system. So if your system is going through time stamps too fast we can't allow level 1 (or 2's) because the time stamp comparison part of the level 1 process (where we are comparing time stamps trying to only pick up pages that changed since the lvl 0 was taken) would brake and your level 1 archive could end up missing pages. So that check was added, and if the system uses time stamps rapidly, then we prevent level 1/2 archives from being taken. I didn't inspect all the code between 11.50.FC5 and 11.50.FC9 but it looks similar. So the first thing I'd throw out there is that your 11.50.FC9 system uses more time stamps then your 11.50.FC5 systems. Jacques Renaut IBM Informix Advanced Support APD Team

First confirm to see if dbspace r1 was actually backed up during the prior level 0 by checking both the online msg log and the onbar activity file I recall that the engine confirms the time stamp of the most recent level 0 from the dbspace root reserved page. So something could be going on in this area where the archive information for the r1 is not getting updated correctly or it's corrupted.

If the page modified timestamp has wrapped since the beginning of the level 0 archive due to high page write activity, then the engine cannot determine with certainly which pages were archived by the preceding level 0 and which were not. That is when this message is produced. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, 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 Wed, Nov 23, 2011 at 9:22 AM, MARK JALKIEWICZ < mark.jalkiewicz@verizon.net> wrote: > First confirm to see if dbspace r1 was actually backed up during the prior > level 0 by checking both the online msg log and the onbar activity file > > I recall that the engine confirms the time stamp of the most recent level 0 > from the dbspace root reserved page. So something could be going on in this > area where the archive information for the r1 is not getting updated > correctly > or it's corrupted. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --0015177406265c52d904b267d422

Jacques, Once again, I believe you have nailed it! This system is extremely busy and goes through timestamps like (pick your favorite analogy). I never gave timestamps a consideration when looking into this. Thanx again, Dan