RE: Any assistance greatly appreciated: [4079]

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_01C50060.00C55540 Content-Type: text/plain All, I am now unable to reply individually to your emails, as I am receiving quite a few. I would like to thank you all, however no solutions have yet been discovered, IBM Support have a wonderful message on their phone line at the moment (well here in Australia and on the number I was given): "Due to circumstances out of our control we are unable to accept your phone call at the moment, please try again later.". That is really not what you want to hear when the powers at be are jumping on you. (why are problems always Friday afternoon ??? :) ...) I have seen this error once before about 10 years or so ago and it was a recoverable state I just cannot remember how it was achieved - I think it was a simple matter of Informix logging in and setting a flag within the root dbspace but in that I cannot get on to IBM it makes it difficult to achieve. I thank you all for your help this far, and will happily receive any and all mails that may offer any sort of hope. When a solution is found ( confidence :) ) I will post the solution so at least it is in the archives so in 10 years when I have the problem again at least it is noted somewhere :). Thanks. Cliff. PS: Is the light at the end of the tunnel an on coming train? that's the question! _____ From: Burton, Clifford [mailto:clifford_burton@simplus.com.au] Sent: Saturday, 22 January 2005 6:06 PM To: Ravi T Cc: informix-list@iiug.org Subject: RE: Any assistance greatly appreciated: [4079] Ravi, Unfortunately the nature of the error isn't attributed to the memory creation (the reference to the single shared memory segment merely states that the allocation of 1.5Gb of memory exceeds the operating systems maximum shared memory allocation and therefore is proceeding to create 2 memory segments ~ there is an overhead to this but that is outside the realms of this thread). The actual problem is that the database instance will not initialise with a message that the "checking database partition index failed" after a cold restore. There is a curiosity with the informix log in so far as it does not record that anything is wrong, it completes the physical and logical restores and there are no more entries in the message log. Thanks for your efforts, they are appreciated. Cheers Cliff. _____ From: Ravi T [mailto:ravit29@yahoo.com] Sent: Saturday, 22 January 2005 5:55 PM To: Burton, Cli.... Subject: Re: Any assistance greatly appreciated: [4079] You can do an ipcs and see if there is any hanging segments, if so then kill them using the command ipcrm -m <segmentnumber>. Or you can use ps -eaf |grep oninit and kill all the oninit processes and then confirm that no segments are hanging using ipcs command. Then, try to bring the system up. Remember, you should use the above commands, only if you have a single instance housed in the box you are woking on. Otherwise you have to map the memory segments to the instances and see that you are killing the correct unused segments only prior to killing them or you will bring down all the instances. Check onstat -m and you will see that the memory segment it is complaining about is listed there and you can compare that segment to the output from ipcs to get a better idea. Hope this helps. "Burton, Cli...." <clifford_burton@simplus.com.au> wrote: Hi All, I have a situation where the IDS-731UD4 instance I run on HP-UX 11 is failing after a cold restore. I do not have time to perform another restore without serious business ramifications. The output from the log file is: the last lines: 09:47:30 Could not create single shared memory segment with resident and non-re sident partitions. Proceeding to create 2 shared memory segments instead. 09:47:32 Segment locked: addr=0xc1841000, size=683679744 Sat Jan 22 09:47:32 2005 09:47:32 Event alarms enabled. ALARMPROG = '/apps/informix_ids/etc/ifx_alert.sh' 09:47:38 DR: DRAUTO is 0 (Off) 09:47:39 Dynamically allocated new message shared memory segment (size 1104KB) 09:47:39 Informix Dynamic Server Version 7.31.UD1 Software Serial Number AAD #J216570 09:47:39 Informix Dynamic Server Initialized -- Shared Memory Initialized. 09:47:39 Physical Recovery Started. 09:47:39 Physical Recovery Complete: 0 Pages Restored. 09:47:39 Logical Recovery Started. 09:47:42 Logical Recovery Complete. 0 Committed, 0 Rolled Back, 0 Open, 0 Bad Locks However: informix@machinename$ onstat -shared memory not initialized for INFORMIXSERVER 'newworld' The results of the oninit -v are: Checking group membership to determine server run modesucceeded Reading configuration file '/apps/informix_ids/etc/onconfig.newworld'...succeeded Creating /INFORMIXTMP/.infxdirs ... succeeded Creating infos file "/apps/informix_ids/etc/.infos.newworld" ... "/apps/informix_ids/etc/.conf.newworld" ... succeeded Writing to infos file ... succeeded Checking config parameters...succeeded Allocating and attaching to shared memory...succeeded Creating resident pool 143252 kbytes...succeeded Creating buffer pool 524290 kbytes...succeeded Initializing rhead structure...succeeded Initializing ASF ...succeeded Initializing Dictionary Cache and Stored Procedure Cache...succeeded Bringing up ADM VP...succeeded Creating VP classes...succeeded Onlining 4 additional cpu vps...succeeded Onlining 18 IO vps...succeeded Forking main_loop thread...succeeded Initialzing DR structures...succeeded Forking 1 'ipcshm' listener threads...succeeded Forking 2 'soctcp' listener threads...succeeded Forking 1 'ipcshm' listener threads...succeeded Forking 0 'soctcp' listener threads...succeeded Forking 0 'soctcp' listener threads...succeeded Starting tracing...succeeded Initializing 32 flushers...succeeded Initializing log/checkpoint information...succeeded Opening primary chunks...succeeded Opening mirror chunks...succeeded Initializing dbspaces...succeeded Validating chunks...succeeded Initialize Async Log Flusher...succeeded Forking btree cleaner...succeeded Initializing DBSPACETEMP list Checking database partition index...FAILED oninit: Fatal error in shared memory initialization I realise that it is unlikely that too many people are out there right at this moment, however, I have been working on this issue so long now that I am loosing track of time, Does anyone have any ideas as to how to get this online, I do have a vague recollection of an issue like this in the past that was resolved quite simply. I am attempting to contact IBM Support on the issue although it is a moderately complex task out of hours when I cannot locate the support contracts etc... Clifford Burton Senior Database Administrator SIMplus Mobile Pty Ltd [ www.SIMplus.com.au ] Ph: (02) 9028 1716 Fx: (02) 9030 6284 Mb: 0414 381 381@@N