Fixing PTF UA95941 (HIPER) is available. We happen to have it on because we (very) recently installed the latest RSU plus FIXCAT items. We never saw the error but will now be especially careful not to propagate the vulnerable level in the enterprise. Thanks again Sam!
. . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW [email protected] -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Jousma, David Sent: Friday, June 29, 2018 7:51 AM To: [email protected] Subject: (External):Re: OA55296 data loss DB2 may be especially vulnerable Thanks Sam! I've opened PMR with IBM. We've been V2.3 since March, and have not seen this issue. I've asked them what the conditions are for the problem to occur. Fixing PTF requires IPL. There is also another related APAR OA55525 that is still open, but the local fix in it is to use ICKDSF to rebuild the VTOX index (from a system that has OA55296 installed). _________________________________________________________________ Dave Jousma Manager Mainframe Engineering, Assistant Vice President [email protected] 1830 East Paris, Grand Rapids, MI 49546 MD RSCB2H p 616.653.8429 f 616.653.2717 -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Knutson, Samuel Sent: Friday, June 29, 2018 7:54 AM To: [email protected] Subject: OA55296 data loss DB2 may be especially vulnerable **CAUTION EXTERNAL EMAIL** **DO NOT open attachments or click on links from unknown senders or unexpected emails** Nofars shared this on Twitter and I don't see it in the last month here on IBM-MAIN so in case others have not seen it. https://twitter.com/nofars/status/1012216001201606657 People, DO NOT UPGRADE TO z/OS 2.3 without OA55296 applies. We just had a few dozen DB2 datasets destroyed. #zOS #mainframe To be more specific- because DB2 datasets tend to have a large amount of extents, that's where you'll likely get hit by this. We are now stuck with inaccessible datasets that can't even be deleted. The VTOC and VTOCIX appear to be beyond repair #mainframe Not sure why this isn't a Red Alert. #mainframe #zOS Ouch! http://www-01.ibm.com/support/docview.wss?uid=isg1OA55296 OA55296: F4 DSCB IN VTOC OVERLAID WITH ZEROS RESULTING IN VOLUME BEING INACCESSIBLE, release 2.3 only A fix is available http://www-01.ibm.com/support/docview.wss?uid=isg1OA55296#more APAR status * Closed as program error. Error description * VTOC CONVERT routine writes F0 DSCB in place of F4 DSCB * resulting in IEC603I and IEC143I 213-04 when attempting to open * the VTOC. This is triggered by an orphaned F3 DSCB in the VTOC. * * Sequence of events: * 1. IEC608I DADSM FUNCTION DISABLED THE VTOC INDEX ON * dddd,vvvvvv,20,CCCCHHHHR * Cde 20 : The Allocate or Extend function discovered that a * DSCB marked unallocated in the VTOC Index, indicating that * the DSCB is a Format 0, is not a Format 0 DSCB. The cchhr is * that of the DSCB. * * 2. DUMP DADSM CREATE (or EXTEND) DIAG=0820040B or DIAG=0831040B * or DIAG=0853040B * The key is that there is a DIAG=xxxx040B, which means CVAF RC * = 04 ; CVSTAT = 0B : DSCB is not format-0 DSCB and VERIFY=YES * * 3. IEC604I VTOC CONVERT ROUTINE ENTERED ON dddd,vvvvvv,,DEVMAN * If the DEVMAN keyword is present, this indicates that * conversion is taking place in the Device Manager Address * Space. * * 4. IEC603I VTOC ERRORS MAY EXIST ON dddd,vvvvvv,0 * Cde 0: An I/O or unexpected CVAF error return code error * occurred in the DADSM routines Allocate, Extend, Scratch or * Partial Release. The routine ended processing, thereby * leaving the VTOC in a partially updated state. * * 5. Attempts to open SYS1.VTOC result in IEC143I 213-04 * Local fix * contact DADSM L2 to recover volume. * . * ++APAR available as follows.... * On server testcase.boulder.ibm.com in the fromibm/mvs * directory in binary format as BA55296.HDZ2230 * Problem summary * **************************************************************** * * USERS AFFECTED: * * * Users of volumes with Indexed VTOC. * * **************************************************************** * * PROBLEM DESCRIPTION: * * * During the process of converting Indexed VTOC to OSVTOC (due * * * to administrator action or error detection), if an orphaned * * * Format-3 DSCB is detected, it will be reset to Format-0 * * * DSCB. The CCHHR address of the orphaned Format-3 DSCB was * * * reset to CCHHR of the Format-4 DSCB, causing that to be the * * * DSCB reset to format-0. This caused the VTOC to be * * * inaccessible. * * **************************************************************** * * RECOMMENDATION: * * **************************************************************** * Problem conclusion * The problem has been corrected to prevent Format-4 DSCB from * being overwritten with Format-0 DSCB. * Temporary fix Comments APAR Information * APAR number OA55296 * Reported component name DEVICE MGMT SER * Reported component ID 5695DF133 * Reported release 230 * Status CLOSED PER * PE NoPE * HIPER YesHIPER * Special Attention NoSpecatt / Xsystem * Submitted date 2018-04-11 * Closed date 2018-04-19 * Last modified date 2018-06-04 * APAR is sysrouted FROM one or more of the following: * APAR is sysrouted TO one or more of the following: UA95941 Modules/Macros * IGG0425P * Fix information * Fixed component name DEVICE MGMT SER * Fixed component ID 5695DF133 Applicable component levels * R230 PSY https://www14.software.ibm.com/webapp/set2/ordermedia/shopCart?ptfs=UA95941 UP18/05/04 P F805 ¢ Fix is available * Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries. Best Regards, Sam Knutson | VP, Product Management | Compuware @samknutson | linkedin.com/in/samknutson [email protected] | M: +1 301 996-1318 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
