Patch-ID# 111948-09 Keywords: SNDR errors SUNWrdcu sndradm bitmap panics reconfig configuration Synopsis: Data Services 3.0 SNDR: sndradm/rdc/sndrd/rdcsrv fixes Patch Date: Jun/17/2003 Install Requirements: Install in Single User Mode Reboot after installation Solaris Release: 2.6 7 8 SunOS Release: 5.6 5.7 5.8 Unbundled Product: StorEdge Instant Image Unbundled Release: 3.0 Xref: Topic: Relevant Architectures: sparc NOTE: The Data Services products contain common core components. The patch ids for these core components are: 111945 111946 It is highly recommended that these patches be installed in addition to any Data Services product patches. NOTE: After installing patch 111948-09, all sets for which autosync is enabled will have to have autosync re-enabled in order for the autosync functionality to operate properly. This should be done once the system has been rebooted after installing the patch. NOTE: This problem exists for non-clustered systems on which SNDR 3.0 / 3.0.1 sets were enabled prior to applying patch 111948-09. After applying patch 111946-05, svadm may produce warning messages like the following for all SV devices associated with the SNDR sets: svadm: warning: /dev/vx/rdsk/dsdg/vol_0 is configured, but not in the config storage The reason for this is that sndradm before adding 111948-09 creates invalid SV entries in the configuration (see bug id 4545796 - sndradm inserts bad sv entries) and patch 111946-05 causes svadm to recognize these entries as invalid. For example: A valid SV entry resembles the following: sv: /dev/vx/rdsk/testdg/50mvol2 - - The invalid entries created by sndradm resemble the following: sv: /dev/vx/rdsk/testdg/50mvol2 - (the last field is missing from the configuration entry) These warning messages are a result of a cosmetic problem only; they are not indicative of problems in functionality. However, to stop the warning messages from appearing, follow the procedure outlined below: 1) Manually create a file that contains the names of all the devices for which invalid SV entries were created. For example, if svadm produces the following output: svadm: warning: /dev/vx/rdsk/dsdg/vol_0 is configured, but not in the config storage svadm: warning: /dev/vx/rdsk/dsdg/vol_1 is configured, but not in the config storage svadm: warning: /dev/vx/rdsk/dsdg/vol_2 is configured, but not in the config storage then create a file that contains the names of those devices: For example: # cat config.txt /dev/vx/rdsk/dsdg/vol_0 /dev/vx/rdsk/dsdg/vol_1 /dev/vx/rdsk/dsdg/vol_2 # 2) Disable those SV devices via the file just created # svadm -df config.txt 3) Reenable the SV devices via the file # svadm -ef config.txt Please note that duplicate entries will then exist in dscfg for the reenabled SV devices, but this will not cause any problems with functionality. The output of svadm will then display the SV devices properly. Patch 111948-09 fixes bug id 4545796. This procedure is only to be executed against SV volumes enabled via sndradm prior to patch 111948-09 Note: After installing 111948-09, the reporting of autosync will be altered. `sndradm -P` will report the active state of autosync. "on" will indicate that autosync is active, "off" will indicate that autosync is not active. After installing this patch, autosync is only active from the time a copy/update is issued on a set(s) until the set(s) are manually put into logging mode. To determine if autosync has been enabled for a set, use `dscfg -l` to verify the autosync value for a set(s). When autosync has been enabled, the tag "auto=on" will be added to the end of the entry in the configuration database for the set(s). If this value is not present, then autosync has not been enabled for the set(s). BugId's fixed with this patch: 4349886 4420589 4429316 4431925 4433407 4452872 4453357 4453470 4459512 4462084 4462376 4463891 4467355 4468096 4468508 4468527 4472717 4478281 4479522 4480148 4484034 4485959 4486324 4487385 4487561 4489165 4490194 4492292 4494630 4510785 4511398 4512334 4516954 4520748 4521899 4523443 4524934 4527038 4527793 4532465 4533711 4545796 4615525 4628594 4631433 4644301 4652076 4653388 4690378 4711396 4747610 4766602 4783254 4799326 4814213 4823492 Changes incorporated in this version: 4747610 4766602 4783254 4799326 4814213 4823492 Patches accumulated and obsoleted by this patch: Patches which conflict with this patch: Patches required with this patch: 111945-09 or greater Obsoleted by: Files included with this patch: /usr/kernel/drv/rdc-5.6 /usr/kernel/drv/rdc-5.7 /usr/kernel/drv/rdc-5.8 /usr/kernel/drv/sparcv9/rdc-5.7 /usr/kernel/drv/sparcv9/rdc-5.8 /usr/kernel/misc/rdcsrv-5.6 /usr/kernel/misc/rdcsrv-5.7 /usr/kernel/misc/rdcsrv-5.8 /usr/kernel/misc/sparcv9/rdcsrv-5.7 /usr/kernel/misc/sparcv9/rdcsrv-5.8 /usr/opt/SUNWesm/SUNWrdc/man/man1rdc/rdcadm.1m /usr/opt/SUNWesm/SUNWrdc/man/man1rdc/sndradm.1m /usr/opt/SUNWesm/SUNWrdc/man/man1rdc/sndrstat.1m /usr/opt/SUNWesm/SUNWrdc/sbin/sndradm /usr/opt/SUNWesm/SUNWrdc/sbin/sndrboot /usr/opt/SUNWesm/SUNWrdc/sbin/sndrstat /usr/opt/SUNWesm/cluster/sbin/rdc /usr/opt/SUNWesm/sbin/rdcadm /usr/opt/SUNWrdc/lib/sndrd /usr/opt/SUNWrdc/lib/sndrsyncd-5.6 /usr/opt/SUNWrdc/lib/sndrsyncd-5.7 /usr/opt/SUNWrdc/lib/sndrsyncd-5.8 Problem Description: 4814213 disabling autosync can be delayed causing un-wanted update syncs 4823492 sndrsyncd dies on cluster 4799326 autosync should resume for set after cluster failover 4783254 putting an SNDR set in a group removes diskgroup data from config file 4766602 sndradm and dscfg output incorrectly report replication mode of SNDR sets 4747610 autosync breaks cluster failover (from 111948-08) 4690378 SNDR- Autosync option doesn't work on E10K 4711396 SNDR- Autosync option is broken (from 111948-07) 4652076 panic after delete finishes on sndr secondary system 4653388 system can deadlock when bitmap reference overflows (from 111948-06) 4644301 BAD TRAP in rdc 4628594 sndradm man page states that set must be in logging mode to reconfig bitmaps (from 111948-05) 4631433 secondary sndr system panics during delete 4615525 Executing enable command on inactive cluster node causes a segmentation fault. 4545796 sndradm inserts bad sv entries into dscfg 4533711 sndradm -m reports wrong error message 4532465 SNDR Async mode hangs the whole system while doing heavy I/O 4527793 no warning, usage, or error when -g groupname is used in sndradm -e command 4527038 sndr writes to volume in an existing sndr set, if virtual hostname is used 4523443 SNDR bitmap gets reconfigured inconsistently 4521899 sndrd can appear to hang after startup 4520748 rdcadm does not perform checking for volumes names over 63 chars when enabling 4516954 sndradm -r -m does not set return error value when primary still mounted 4512334 sndradm -p doesn't work for non-root users 4510785 During a full reverse sync isnt checked to see if it is mounted. 4452872 SNDR reconfiguration of bitmap volume with -f option does not work (from 111948-04) 4524934 Data Services -04 patches no longer recognize with local Suncluster3.0_U1 mode 4511398 sndradm can create duplicate sv entries 4494630 sndradm -I doesn't check validity of volumes used in ndr_ii set, nor does II 4489165 sndradm core dumps on a clustered node which is not mastering the diskgroup 4484034 if ds config location is unwritable,config is still entered in kernel (from 111948-03) 4492292 sndrstat -v does not print verbose information. 4487385 sndradm manpages does not specify return value / exit status (from 111948-02) 4490194 Secondary cluster node panics on reboot after forceing a core dump 4487561 Complex multihomed configurations with one to many will fail 4486324 SNDR secondary panics when sndradm -d is issued 4462084 rdcadm -R b s allows volumes in use to be used as a bitmap 4349886 sndr3.0 bitmaps can be used on top of master devices (from 111948-01) 4485959 sndradm hard link is broken as a result of 3.0 SNDR patch 4480148 Async flushing thread may become stuck 4479522 system panics with assertion failure 4478281 sndrd error messages wrong 4472717 Putting volumes in logging mode on secondary VERY slow while primary is down 4468527 sndradm -d when more than 2 SNDR vols setup results in hang on dscfg write lock 4468508 sndradm -d on secondary hangs w/ deadlock rdc_conf_lock 4468096 sndradm is returning error trying to operate on device group which is on another 4467355 sndrstat -h flag displays usage info for rdcstat command 4463891 sndrboot fails to resume local volume 4462376 SNDR bitmap conversion procedure does not work for remote host 4459512 iiadm -e does not return propt w/sndr volume used as master until sndr is comple 4453470 rdcadm command allows block devices in its rdc-set 4453357 SNDR - Autosync option doesn't work 4433407 Switchover of disk set may cause sndr to become confused about its identity 4431925 sndr 3.0 enabled on the primary server first, message that interface is down 4429316 Cache cannot disable if sndr bitmap volumes are in use 4420589 error message outputs incorrectly Patch Installation Instructions: -------------------------------- For Solaris 2.0 - 2.6 releases, refer to the Install.info file and/or the README within the patch for instructions on using the generic 'installpatch' and 'backoutpatch' scripts provided with each patch. For Solaris 7-8 releases, refer to the man pages for instructions on using 'patchadd' and 'patchrm' scripts provided with Solaris. Any other special or non-generic installation instructions should be described below as special instructions. The following example installs a patch to a standalone machine: example# patchadd /var/spool/patch/104945-02 The following example removes a patch from a standalone system: example# patchrm 104945-02 For additional examples please see the appropriate man pages. Special Install Instructions: ----------------------------- Since this patch updates modules that live in the kernel, it is necessary for the user to boot the system up in single user mode to apply the patch and then reboot the system. README -- Last modified date: Tuesday, June 17, 2003