10-13-2015 02:24 PM
Hello ppl,
I need quick help to understand the below situation as i am a bit new to VCS world .
We have a global cluster running with 2 nodes on each site . Primary side (2) and secondary site 2.
The secondary site however is complete idle doing nothing and just gettng data replicated via storage disk groups.
There are usually 2 disk group on each site cluster (ossdg and sybasedg) when active . From few days i am observing on secondary site nodes (admin1 and 2) Vxvm disk errors and its hard for me to figure out is it nomral beacuse it si idle cluster. The disks on storgare looks fine with no failure however Vxvm gives error and OS doesnt see it too . Also i dont see any output for vxprint . Is this normal for global cluster. Please help me figure out if i need to take an action .
PFB Vxdisk list ouput :
Seondary cluster server 1
[23:18:11 root@xxx-xyz-admin1 /]>> vxdisk list
DEVICE TYPE DISK GROUP STATUS
c0t0d0s2 auto:SVM - - SVM
c0t1d0s2 auto:SVM - - SVM
c1t500601623DE03785d0s2 auto:sliced c1t500601623DE03785d0s2 vxfendg1 on line
c1t500601623DE03785d1s2 auto:sliced c1t500601623DE03785d1s2 vxfendg1 on line
c1t500601623DE03785d2s2 auto:sliced c1t500601623DE03785d2s2 vxfendg1 on line
c1t500601623DE03785d3s2 auto:sliced - - online
c1t500601623DE03785d4s2 auto:sliced - - online
c1t500601623DE03785d5s2 auto:sliced - - online
c1t500601623DE03785d6s2 auto - - error
c1t500601623DE03785d7s2 auto - - error
c1t500601623DE03785d8s2 auto - - error
secondary cluster server 2
[23:19:40 root@xxx-xyz-admin2 /]>> vxdisk list
DEVICE TYPE DISK GROUP STATUS
c0t0d0s2 auto:SVM - - SVM
c0t1d0s2 auto:SVM - - SVM
c1t500601623DE03785d0s2 auto:sliced - - online
c1t500601623DE03785d1s2 auto:sliced - - online
c1t500601623DE03785d2s2 auto:sliced - - online
c1t500601623DE03785d3s2 auto:sliced - - online
c1t500601623DE03785d4s2 auto:sliced - - online
c1t500601623DE03785d5s2 auto:sliced - - online
c1t500601623DE03785d6s2 auto - - error
c1t500601623DE03785d7s2 auto - - error
c1t500601623DE03785d8s2 auto - - error
c1t500601623DE03785d9s2 auto - - error
Server icluster runs in active /active mode.
Let me know if any extrax info is required to analyse this.
Solved! Go to Solution.
10-23-2015 03:19 AM
HI,
As mentioned before that EMC Mirrorview LUNS are seen in error state on the secondary site in VxVM, this is becasue the replication technology does not provide access to the device metadata on the secondary site.
You can refer to the below URL
https://www.veritas.com/support/en_US/article.HOWTO32485
https://www.veritas.com/support/en_US/article.HOWTO32495
Regards,
Sudhir
10-15-2015 06:05 AM
Hi ,
I am still loking for some help
Rgds
Mohammed
10-16-2015 12:50 AM
Which array based replication are you using?
Regards,
Sudhir
10-16-2015 12:51 AM
Such a behaviour is observed in case of EMC Mirrorview technology.
Regards,
Sudhir
10-16-2015 12:57 AM
Hello,
about vxprint, you are not getting outputs on secondary because diskgroup is deported ... so that is normal
secondly, this doesn't looks to be a A/A cluster, as your secondary site is idle, it will takeover if some failure happens or someone forces secondary to be active .
Now about the errors you are getting,
1. Paste the exact error messages here
2. What is the version of OS & VxVM ?
3. I am assuming your replication is VVR here ? or something else ?
4. Have you made sure that disks which are giving errors on secondary are indeed part of the replicated data set from primary ? are there any additional disks in system which are not participating in replication ?
G
10-16-2015 01:01 AM
Also, I see that this post was posted in "Storage & clustering documentation", I have moved it to "Storage Foundation" forum so that others can see & contribute
G
10-16-2015 11:05 AM
Hello ,
As Sudhir Mentioned we are using EMC Mirrorview technology and not VVR.
ok i agree for vxprint not showing any results. I am not fully sure if the disk with errors is a part of data replication set , but i assume yes if i see the disk will typically form critical disk groups when cluster will be made online
OS Version :
uname -a
SunOS 5.10 Generic_150401-10 i86pc i386 i86pc
VxVM version :
PKGINST: VRTSvxvm
NAME: Binaries for VERITAS Volume Manager by Symantec
CATEGORY: system
ARCH: i386
VERSION: 5.1,REV=10.06.2009.21.48
BASEDIR: /
VENDOR: Symantec Corporation
DESC: Virtual Disk Subsystem
PSTAMP: 5.1.103.000-5.1SP1RP3_x86-2012-09-25-142630-16
INSTDATE: Mar 04 2014 20:02
HOTLINE: http://support.veritas.com/phonesup/phonesup_ddProduct_.htm
EMAIL: support@veritas.com
STATUS: completely installed
FILES: 876 installed pathnames
44 shared pathnames
113 directories
402 executables
376060 blocks used (approx)
What more error should i paste ?
One extra info i can view the device from vxddl , below marked bold
> vxddladm list devices
DEVICE TARGET-ID STATE DDL-STATUS (ASL)
===============================================================
c0t1d0s2 - Online CLAIMED (Disk)
c0t0d0s2 - Online CLAIMED (Disk)
c1t500601693DE03785d0s2 c1_p4_t0 Online -
c1t500601693DE03785d1s2 c1_p4_t0 Online -
c1t500601693DE03785d2s2 c1_p4_t0 Online -
c1t500601693DE03785d3s2 c1_p4_t0 Online -
c1t500601693DE03785d4s2 c1_p4_t0 Online -
c1t500601693DE03785d5s2 c1_p4_t0 Online -
c1t500601693DE03785d6s2 c1_p4_t0 Online -
c1t500601693DE03785d7s2 c1_p4_t0 Online -
c1t500601693DE03785d8s2 c1_p4_t0 Online -
c1t500601693DE03785d9s2 c1_p4_t0 Online -
c1t500601623DE03785d0s2 c1_p2_t0 Online -
c1t500601623DE03785d1s2 c1_p2_t0 Online -
c1t500601623DE03785d2s2 c1_p2_t0 Online -
c1t500601623DE03785d3s2 c1_p2_t0 Online -
c1t500601623DE03785d4s2 c1_p2_t0 Online -
c1t500601623DE03785d5s2 c1_p2_t0 Online -
c1t500601623DE03785d6s2 c1_p2_t0 Online CLAIMED (libvxCLARiiON.so)
c1t500601623DE03785d7s2 c1_p2_t0 Online -
c1t500601623DE03785d8s2 c1_p2_t0 Online -
c1t500601623DE03785d9s2 c1_p2_t0 Online -
c2t5006016A3DE03785d0s2 c2_p4_t0 Online -
c2t5006016A3DE03785d1s2 c2_p4_t0 Online -
c2t5006016A3DE03785d2s2 c2_p4_t0 Online -
c2t5006016A3DE03785d3s2 c2_p4_t0 Online -
c2t5006016A3DE03785d4s2 c2_p4_t0 Online -
c2t5006016A3DE03785d5s2 c2_p4_t0 Online -
c2t5006016A3DE03785d6s2 c2_p4_t0 Online -
c2t5006016A3DE03785d7s2 c2_p4_t0 Online -
c2t5006016A3DE03785d8s2 c2_p4_t0 Online -
c2t5006016A3DE03785d9s2 c2_p4_t0 Online -
c2t500601613DE03785d0s2 c2_p2_t0 Online -
c2t500601613DE03785d1s2 c2_p2_t0 Online -
c2t500601613DE03785d2s2 c2_p2_t0 Online -
c2t500601613DE03785d3s2 c2_p2_t0 Online -
c2t500601613DE03785d4s2 c2_p2_t0 Online -
c2t500601613DE03785d5s2 c2_p2_t0 Online -
c2t500601613DE03785d6s2 c2_p2_t0 Online -
c2t500601613DE03785d7s2 c2_p2_t0 Online -
c2t500601613DE03785d8s2 c2_p2_t0 Online -
c2t500601613DE03785d9s2 c2_p2_t0 Online -
Regards,
Mohammed
10-18-2015 10:16 PM
Hi,
If you are using EMC Mirroeview than the disk son the secondary site show as error in VxVM. The diskgroup will be imported correctly when the replication role is changed from secondary to primar when you try to bring the service group online on the DR site.
Were these disk in a different state before? Do u have anyother cluster with similar configurations showing a different disk state?
Regards,
Sudhir
10-18-2015 11:08 PM
What is the version of ASLAPM in use ?
# pkginfo -l VRTSaslapm
seems like linux had some issues with lvm headers which was fixed in hotfix, worth to find if something similar in Solaris was discovered
https://www.veritas.com/support/en_US/article.TECH179423
G
10-19-2015 12:07 PM
Hi ,
To Answer @Sudhir : I am really not sure as i have been given responsibility of this recently ,and this is the only global cluster setup we have.
@ Gaurav :
pkginfo -l VRTSaslapm
PKGINST: VRTSaslapm
NAME: Array Support Libraries and Array Policy Modules for Veritas Volume Manager.
CATEGORY: system
ARCH: i386
VERSION: 5.1.100.400,REV=04.14.2012.07.08
BASEDIR: /
VENDOR: Symantec Corporation
DESC: Virtual Disk Subsystem
PSTAMP: 5.1.100.400,2012-04-14
INSTDATE: Mar 04 2014 18:36
HOTLINE: http://www.symantec.com/business/support/assistance_care.jsp
STATUS: completely installed
FILES: 225 installed pathnames
20 shared pathnames
20 directories
121 executables
6685 blocks used (approx)
You might be true with the "linux article" but what surprises me is that the disks cannot be seen at OS layer too . PFB .
21:04:14 root@xyz /]>> echo | format | grep -i dri
8. c1t500601623DE03785d6 <drive type unknown>
9. c1t500601623DE03785d7 <drive type unknown>
10. c1t500601623DE03785d8 <drive type unknown>
11. c1t500601623DE03785d9 <drive type unknown>
18. c1t500601693DE03785d6 <drive type unknown>
19. c1t500601693DE03785d7 <drive type unknown>
20. c1t500601693DE03785d8 <drive type unknown>
21. c1t500601693DE03785d9 <drive type unknown>
28. c2t5006016A3DE03785d6 <drive type unknown>
29. c2t5006016A3DE03785d7 <drive type unknown>
30. c2t5006016A3DE03785d8 <drive type unknown>
31. c2t5006016A3DE03785d9 <drive type unknown>
38. c2t500601613DE03785d6 <drive type unknown>
39. c2t500601613DE03785d7 <drive type unknown>
40. c2t500601613DE03785d8 <drive type unknown>
41. c2t500601613DE03785d9 <drive type unknown>
[21:04:38 root@xyz /]>> vxdisk list
DEVICE TYPE DISK GROUP STATUS
c0t0d0s2 auto:SVM - - SVM
c0t1d0s2 auto:SVM - - SVM
c1t500601623DE03785d0s2 auto:sliced c1t500601623DE03785d0s2 vxfendg1 on line
c1t500601623DE03785d1s2 auto:sliced c1t500601623DE03785d1s2 vxfendg1 on line
c1t500601623DE03785d2s2 auto:sliced c1t500601623DE03785d2s2 vxfendg1 on line
c1t500601623DE03785d3s2 auto:sliced - - online
c1t500601623DE03785d4s2 auto:sliced - - online
c1t500601623DE03785d5s2 auto:sliced - - online
c1t500601623DE03785d6s2 auto - - error
c1t500601623DE03785d7s2 auto - - error
c1t500601623DE03785d8s2 auto - - error
c1t500601623DE03785d9s2 auto - - error
Rgds,
Mohammed
10-23-2015 03:19 AM
HI,
As mentioned before that EMC Mirrorview LUNS are seen in error state on the secondary site in VxVM, this is becasue the replication technology does not provide access to the device metadata on the secondary site.
You can refer to the below URL
https://www.veritas.com/support/en_US/article.HOWTO32485
https://www.veritas.com/support/en_US/article.HOWTO32495
Regards,
Sudhir