Patch-ID# 113605-02 Keywords: vxfs 4.0 veritas file system rolling patch 02 Synopsis: VERITAS File System 4.0: VxFS 4.0 Rolling Patch 02 Date: Jun/15/2004 ********************************************************** Note: Please download the patch 116688-xx instead of this patch 113605. All the fixes in 113605 are included in 116688-xx. ********************************************************** Install Requirements: Reboot after installation Solaris Release: 8 SunOS Release: 5.8 Unbundled Product: Veritas VxFS Unbundled Release: 4.0 Xref: Topic: VxFS 4.0 RP2 Multiple fixes patch Relevant Architectures: sparc BugId's fixed with this patch: 4667568 Changes incorporated in this version: 4667568 Patches accumulated and obsoleted by this patch: Patches which conflict with this patch: Patches required with this patch: Obsoleted by: Files included with this patch: /etc/fs/vxfs/mount /etc/fs/vxfs/qlogattach /etc/fs/vxfs/qlogck /etc/fs/vxfs/qlogrec /kernel/drv/sparcv9/vxportal /kernel/drv/vxportal /kernel/fs/sparcv9/vxfs /kernel/fs/vxfs /opt/VRTSvxfs/lib/libvxfsutil.a /opt/VRTSvxfs/lib/libxdsm.a /opt/VRTSvxfs/lib/sparcv9/libvxfsutil.a /opt/VRTSvxfs/lib/sparcv9/vxckptpriv.so /opt/VRTSvxfs/lib/sparcv9/vxfspriv.so /opt/VRTSvxfs/lib/sparcv9/vxfsutil.so /opt/VRTSvxfs/lib/vxckptpriv.so /opt/VRTSvxfs/lib/vxfspriv.so /opt/VRTSvxfs/lib/vxfsutil.so /opt/VRTSvxfs/sbin/cp /opt/VRTSvxfs/sbin/cpio /opt/VRTSvxfs/sbin/fcladm /opt/VRTSvxfs/sbin/fsadm /opt/VRTSvxfs/sbin/fsapadm /opt/VRTSvxfs/sbin/fscat /opt/VRTSvxfs/sbin/fscdsadm /opt/VRTSvxfs/sbin/fscdsconv /opt/VRTSvxfs/sbin/fsckpt_restore /opt/VRTSvxfs/sbin/fsckptadm /opt/VRTSvxfs/sbin/fsclustadm /opt/VRTSvxfs/sbin/fsenvadm /opt/VRTSvxfs/sbin/fsvoladm /opt/VRTSvxfs/sbin/getext /opt/VRTSvxfs/sbin/ls /opt/VRTSvxfs/sbin/mv /opt/VRTSvxfs/sbin/qioadmin /opt/VRTSvxfs/sbin/qiomkfile /opt/VRTSvxfs/sbin/qiostat /opt/VRTSvxfs/sbin/qlogadm /opt/VRTSvxfs/sbin/qlogckd /opt/VRTSvxfs/sbin/qlogclustadm /opt/VRTSvxfs/sbin/qlogdb /opt/VRTSvxfs/sbin/qlogdetach /opt/VRTSvxfs/sbin/qlogdisable /opt/VRTSvxfs/sbin/qlogenable /opt/VRTSvxfs/sbin/qlogmk /opt/VRTSvxfs/sbin/qlogprint /opt/VRTSvxfs/sbin/qlogrm /opt/VRTSvxfs/sbin/qlogstat /opt/VRTSvxfs/sbin/qlogtrace /opt/VRTSvxfs/sbin/setext /opt/VRTSvxfs/sbin/vxdump /opt/VRTSvxfs/sbin/vxedquota /opt/VRTSvxfs/sbin/vxenablef /opt/VRTSvxfs/sbin/vxfsckd /opt/VRTSvxfs/sbin/vxfsconvert /opt/VRTSvxfs/sbin/vxfsstat /opt/VRTSvxfs/sbin/vxlsino /opt/VRTSvxfs/sbin/vxquot /opt/VRTSvxfs/sbin/vxquota /opt/VRTSvxfs/sbin/vxquotaoff /opt/VRTSvxfs/sbin/vxquotaon /opt/VRTSvxfs/sbin/vxrepquota /opt/VRTSvxfs/sbin/vxrestore /opt/VRTSvxfs/sbin/vxtunefs /opt/VRTSvxfs/sbin/vxupgrade /opt/VRTSvxms/lib/map/libvxfs.so /opt/VRTSvxms/lib/map/sparcv9/libvxfs.so /usr/kernel/drv/fdd /usr/kernel/drv/qlog /usr/kernel/drv/sparcv9/fdd /usr/kernel/drv/sparcv9/qlog /usr/lib/fs/vxfs/df /usr/lib/fs/vxfs/ff /usr/lib/fs/vxfs/fsck /usr/lib/fs/vxfs/fsdb /usr/lib/fs/vxfs/fstyp /usr/lib/fs/vxfs/mkfs /usr/lib/fs/vxfs/ncheck /usr/lib/fs/vxfs/sparcv7/fsck /usr/lib/fs/vxfs/sparcv9/fsck /usr/lib/fs/vxfs/umount Problem Description: -------------------- 4667568 (93459) QIO functionality broken on Solaris 8 above certain Solaris kernel patches. SUN introduced a flag in AIO/KIO and depending upon that flag it determines if pages are locked/unlocked. As VxFS/QIO doesn't set/unset accordingly, Solaris treats them as unlocked pages and this results in hang, etc. (135668) When a file system has clones and is almost full, it could be marked for full fsck while full fsck could not find any defect. (137488) In fsadm, the "maximum time to run" can be reset when running both directory and extent reorg with the same command. (140326) Files/directories could be removed from a directory which had sticky bit set. (141302) Fsck could not replay the intent log sometimes. It sometimes incorrectly determined the head and tail of the log area. (141372) Running "fsck -y" on a corrupted filesystem where the stateino (one of the structural inodes) has been zeroed out was making fsck exit with an error. fsck has now been fixed to not exit but continue and correct the corrupted FS. (141630) VxFS has been updated to log a message when hitting the out of disk space condition, and to fail checkpoint creation if the file system has been marked full fsck required. (142174) Files can become corrupted in cluster file system when multiple parallel writes on the same file are being done. (142986) A panic can occur cluster file system when running two parallel "mkfs" to make VxFS file system on a shared volume. (143127) Certain long-term operations (such as a remove of a large file) result in marking of a file with a state. In the case of a crash (or change of the primary node in a cluster) all of the files with pending states will be processed before the filesystem is again available for the normal use. A file with a pending state after this point is interpreted as a corruption and is marked bad. This change fixes a problem where a certain sequence of operations resulted in not processing the files with pending states. (145930) System crashed in vx_pageio1() when using swap files in VxFS. This problem has been corrected. (from 113605-01) (139587) The primary node of a cluster filesystem may panic after upgrading filesystem layout to version 6 in clustered file system environment. Patch Installation Instructions: -------------------------------- For the Solaris 8 release, refer to the online manual 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/110434-07 The following example removes a patch from a standalone system: example# patchrm 104945-02 For additional examples please see the appropriate manual pages. Special Install Instructions: ----------------------------- You need to use the shutdown command to reboot the system after patch installation or de-installation: shutdown -g0 -y -i6 Additional Notes: ----------------- 1) VERITAS Incident 93459 (Sun BUG ID 4667568) QIO functionality broken on Solaris 8 above certain Solaris kernel patches. SUN introduced a flag in AIO/KIO and depending upon that flag it determines if pages are locked/unlocked. As VxFS/QIO doesn't set/unset accordingly, Solaris treats them as unlocked pages and this results in hang, etc. 2) VERITAS Incident 135668 When a file system has clones and is almost full, it could be marked for full fsck during inactive processing of inodes while full fsck could not find any defect. 3) VERITAS Incident 137488 In fsadm, the "maximum time to run" can be reset when running both directory and extent reorg with the same command. If the directory reorg finishes before the timer expires, fsadm will reset the alarm resulting in longer than expected running times. 4) VERITAS Incident 140326 Files/directories could be removed from a directory which had sticky bit set. Updated VxFS to honor sticky bit on a directory. 5) VERITAS Incident 141302 Fsck could not replay the intent log sometimes. It sometimes incorrectly determined the head and tail of the log area. 6) VERITAS Incident 141372 Running "fsck -y" on a corrupted filesystem where the stateino (one of the structural inodes) has been zeroed out was making fsck exit with an error. fsck has now been fixed to not exit but continue and correct the corrupted FS. 7) VERITAS Incident 141630 VxFS has been updated to log a message when hitting the out of disk space condition, and to fail checkpoint creation if the file system has been marked full fsck required. 8) VERITAS Incident 142174 When multiple parallel writes on the same file are being done in a cluster file system and one of them is extending write, the file can become corrupted. 9) VERITAS Incident 142986 A panic can occur cluster file system when running two parallel "mkfs" to make VxFS file system on a shared volume. 10) VERITAS Incident 143127 Certain long-term operations (such as a remove of a large file) result in marking of a file with a state. In the case of a crash (or change of the primary node in a cluster) all of the files with pending states will be processed before the filesystem is again available for the normal use. A file with a pending state after this point is interpreted as a corruption and is marked bad. This change fixes a problem where a certain sequence of operations resulted in not processing the files with pending states. 11) VERITAS Incident 145930 System crashed in vx_pageio1() when using swap files in VxFS. This problem has been corrected. (from 113605-01) 1) VERITAS Incident 139587 The primary node of a cluster filesystem may panic while migrating the primaryship to another node in the cluster if the filesystem was upgraded from version 5 layout to version 6 layout. README -- Last modified date: Wednesday, December 22, 2004