Unable to boot solaris after un-encapsulating root disk
we have a Solaris 10 X86 "147441-12 " with vxvm version 5.1 installed on a X6270 M2 sun server , The customer have HW raid "RAID 1" configured .. 4 disks configured giving 2 disks to the O.S. these 2 disks are encapsulated using vxvm for rootdisk and root mirror ! After shutting down the server to replace one of the failed disks "HDD2/part of root mirror" , server boot process reported "unexpected free inode. please invoke fsck" .. I booted in failsafe mode , reverted vfstab to vfstab.prevm ... removed rootdev and vxvi lines in /etc/system , created install-db , removed root-done , replaced /a/boot/solaris/bootenv.rc with /a/boot/solaris/bootenv.rc.pre_vxvm .. updated boot archive , reboot into failsafe , did fsck for root disk .. the system hangs while booting with "WARNING: cannot find /etc/svc/volatile" .. if i revert all changes back to vmvm encapsulation it gives the same error .. booting from root mirror gives same error + fsck errors . Please advise ,1.4KViews0likes1CommentNo complete snapshot mirrors are available
# vxsnap -g pri_cluster_resource1 rmmir DB ends up with "No complete snapshot mirrors are available" vxprint shows: [root@sclodb2 ~]# vxprint Disk group: my_cluster_resource TY NAME ASSOC KSTATE LENGTH PLOFFS STATE TUTIL0 PUTIL0 dg my_cluster_resource my_cluster_resource - - - - - - dm backup-a emcpowerg - 2147310624 - - - - dm backup-b emcpowerd - 2147310624 - - - - dm db-a emcpowerh - 2147310624 - - - - dm db-b emcpowers - 2147310624 - - - - dm jrnalt-a emcpowere - 209534256 - - - - dm jrnalt-b emcpowerm - 209534256 - - - - dm jrnpri-a emcpowerj - 209534256 - - - - dm jrnpri-b emcpowerq - 209534256 - - - - dm log-a emcpowerf - 2147310624 - - - - dm log-b emcpowern - 2147310624 - - - - pl DB_NBU_MIRROR_0-01 - DISABLED 1048576000 - - - - sd backup-b-01 DB_NBU_MIRROR_0-01 ENABLED 1048576000 0 - - - v DB fsgen ENABLED 1048576000 - ACTIVE - - pl DB-01 PMSDBVOLDB ENABLED 1048576000 - ACTIVE - - sd db-a-01 DB-01 ENABLED 1048576000 0 - - - pl DB-02 DB ENABLED 1048576000 - ACTIVE - - sd pmsbi-a-02 DB-02 ENABLED 1048576000 0 - - - dc DB_dco DB - - - - - - v DB_dcl gen ENABLED 72640 - ACTIVE - - pl DB_dcl-01 DB_dcl ENABLED 72640 - ACTIVE - - sd backup-a-01 DB_dcl-01 ENABLED 72640 0 - - - pl DB_dcl-02 DB_dcl ENABLED 72640 - ACTIVE - - sd jrnalt-b-01 DB_dcl-02 ENABLED 72640 0 - - - sp DB_cpmap DB - - - - - - * [root@ ~]# * I have more volumes in my configuration, but they are all fine so I deleted them from the vxprint output. Can somebody help me, to get rid of DB_NBU_MIRROR_0-1, the dc object (vxsnap unprepare not working currently, cause volume is in a snapshot chain) and every entry below DB_dco? Thank you for your help.Solved602Views0likes1CommentTop Technotes Aug/Sept 2009 - Storage Foundation
Your question may already be answered... Here are the recent most-accessed Technotes from Tech Support. For more info, try checking out the Knowledge Base. Storage Foundation for Unix/Linux Late Breaking News (LBN) - Latest additions to the Release Notes for Veritas Storage Foundation and High Availability 5.0, 5.0 Maintenance Pack 1 (MP1), and Maintenance Pack 3 (MP3) for Solaris (SPARC) Hardware Compatibility List for Veritas Storage Foundation (tm) and High Availability Solutions 5.0 Maintenance Pack 1, 5.0 Maintenance Pack 2 (including SFCFS for Oracle RAC for RHEL, SLES and OEL) and 5.0 Maintenance Pack 3 Storage Foundation and High Availability 5.0 MP3 Product Family Documentation Late Breaking News (LBN) - Updates to the Release Notes for Veritas Storage Foundation (tm) and High Availability Solutions 5.0 to 5.0 Maintenance Pack 3 (MP3) on Linux and cross-references to product documentation Veritas Storage Foundation (tm) and High Availability Solutions 5.0 Maintenance Pack 3 How to Check Interface Duplex and Speed Using ndd/kstat on Solaris. Storage Foundation and High Availability 5.0 Product Family Documentation Late Breaking News (LBN) - Updates to the Release Notes for Veritas Storage Foundation and High Availability Solutions 5.0 to 5.0 MP3 products on AIX Veritas Storage Foundation HA 5.0 MP3 RP2a for Solaris Sparc and x86_64 Product Documentation for Storage Foundation and High Availability Solutions 5.0 for Solaris (SPARC) Where can customers download product patches and get new patch availability notifications for Veritas Storage Foundation and High Availability Solutions, Cluster Server One, Cluster Server for VMware ESX, Scalable File Server, and CommandCenteral Storage? After installing or upgrading to Storage Foundation Suite 5.0 (tm), vxsvc fails to start with an error message that the server is not configured. "Corrupt label; wrong magic number" message appears in the Solaris syslog (/var/adm/messages) Dtrace probes cause error message "fbt: WARNING: couldn't allocate FBT table for module vxio" to appear in system logs Volume Manager cannot initialize a disk with no backup slice or a disk with an EFI label Storage Foundation Cluster File System Problem creating a shared disk group after upgrading from VERITAS Storage Foundation (tm) Cluster File System 3.5 or previous version to 4.0. Error message is " Disk group version doesn't support feature; see the vxdg upgrade command" How to resize a CFS file system from any node within the cluster? Unable to online cvm_clus service group; vxfsckd resource will not online Configuring CVMDiskGroup resource type to show: CVMActivation @ = sw is undocumented VERITAS Cluster File Solutions (tm) 4.0 for Linux - an Introduction The AIX snap dump is provided in dump.BZ format after AIX 5.3 Technology Level 5. Directory move operations on secondary mounts in a VERITAS Clustered File System (tm) environment running 3.5 MP2 can cause a system hang. VERITAS Cluster Server resources for VERITAS Cluster File System agents become unavailable Storage Foundation for Databases (Oracle, DB2, Sybase) VCS CRITICAL V-16-1 -50086 CPU usage on is 100% (Oracle) Handling of RAW files with ODM enabled for Oracle version 10g and above (Oracle) How to extend a Quick I/O file (Oracle) How to enable and disable Cached Quick I/O for a file system. (Oracle) How to disable the Quick I/O feature on a file system (Oracle) Is Quick I/O supported in Solaris zones? (DB2) What is the latest version of Veritas Storage Foundation (tm) for DB2 on Linux? (DB2) The vxliccheck utility reports an invalid license for VERITAS Database Edition for Sybase even though a valid license exists. (Sybase) The following error was reported in the Sybase log file when the database tried to write to certain database files: "server bufwritedes: write error detected... server Checkpoint process detected hardware error writing logical page" (Sybase) Converting master or temp datafile to Quick I/O reports: kernel sddone: I/O error Hardware error detected writing logical page '668', virtual page '4768' for dbid '2', cache 'default data cache'. (Sybase)459Views1like0Comments