Setting up Solaris 9 RAID 1. Second hd can't boot after sync
Hi,
I'm trying to setup RAID 1 with Solaris 9. I've got the mirroring working with metadb, metinit and metattach
prtvtoc /dev/rdsk/c1t0d0s2 | fmthard -s - /dev/rdsk/c1t1d0s2
metastat shows both hard disks are sycn and mounting the second hard disk verifies that everything is there
installboot /usr/platform/`uname -i`/lib/fs/ufs/bootblk /dev/rdsk/c1t1d0s0
ls -l /dev/rdsk/c1t1d0s0
at OBP
nvalias disk1 /pci@1f,0/pci@1/scsi@8/sd@
setenv boot-device disk disk1
nvstore
boot disk1
and it gives me the error, "Can't open boot device" and goes back to booting up from disk
Any help? thanks
I'm trying to setup RAID 1 with Solaris 9. I've got the mirroring working with metadb, metinit and metattach
prtvtoc /dev/rdsk/c1t0d0s2 | fmthard -s - /dev/rdsk/c1t1d0s2
metastat shows both hard disks are sycn and mounting the second hard disk verifies that everything is there
installboot /usr/platform/`uname -i`/lib/fs/ufs/bootblk /dev/rdsk/c1t1d0s0
ls -l /dev/rdsk/c1t1d0s0
at OBP
nvalias disk1 /pci@1f,0/pci@1/scsi@8/sd@
setenv boot-device disk disk1
nvstore
boot disk1
and it gives me the error, "Can't open boot device" and goes back to booting up from disk
Any help? thanks
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
The reason it is booting into singleuser mode is that you only have half of the metadb's available. For security reasons it will only boot into singleuser mode then.
In order to get up to multiuser again, remove the replicas of the metadb that are no longer existing (the ones on the failed drive) using metadb -d
It will boot to multiuser if you have at least one more than half of the replicas of the metadb (which you achieve by removing the failed ones)
In order to get up to multiuser again, remove the replicas of the metadb that are no longer existing (the ones on the failed drive) using metadb -d
It will boot to multiuser if you have at least one more than half of the replicas of the metadb (which you achieve by removing the failed ones)
ASKER
I removed the failed replicas and it allowed me to boot up. but after i put back in the other hard disk (both working discs working) in now, it gives me erros
WARNING: /pci@1f,0/pci@1/scsi@8 (glm0):
Connected command timeout for Target 1.0
WARNING: /pci@1f,0/pci@1/scsi@8 (glm0):
Target 1 reducing sync. transfer rate
WARNING: /pci@1f,0/pci@1/scsi@8 (glm0):
got SCSI bus reset
and now i cannot add back the original metadb
WARNING: /pci@1f,0/pci@1/scsi@8 (glm0):
Connected command timeout for Target 1.0
WARNING: /pci@1f,0/pci@1/scsi@8 (glm0):
Target 1 reducing sync. transfer rate
WARNING: /pci@1f,0/pci@1/scsi@8 (glm0):
got SCSI bus reset
and now i cannot add back the original metadb
are you sure you connected it correctly (especially concerning the SCSI-Termination)? These Error Messages sound like bad cabling or wrong termination.
ASKER
i managed to force it to boot with the other disk and added the metadb. but now when i boot up and i do a metastat and it shows the seond submirror needs maintenance
how do i get them to sync in this case, sycnchronising with the second(replacement harddisk)?
how do i get them to sync in this case, sycnchronising with the second(replacement harddisk)?
use metasync on the mirror device.
ASKER
I tried metadsync and it didn't work, so I used metareplace instead
Thank you very much for your help.
ASKER
I went into that folder and it shows everything starts with sd. Anyway i went ahead and changed it to disk and its working now.
But after i booted up, (i removed the first hard disk for testing, leaving it to boot up in the second hard disk) and it gives me error that the d10,d11,d13... (mirror for first hard disk) is missing and boots into s mode instead. how do i remove or make it go offline until i can replace a new hard disk in again?