- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- Re: HW path changed, dsf changed, VGs could be act...
Categories
Company
Local Language
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Community
Resources
Forums
Blogs
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-16-2013 08:09 AM
тАО06-16-2013 08:09 AM
HW path changed, dsf changed, VGs could be activated!
Guys,
What I met early today was beyond my understanding of LVM or io sub-system of HP-UX.
superdome servers with HP-UX 11.11, connected to SAN switches to access EMC storage, assuming that dsf of LUNs in EMC storage were c21t*d* & c22t*d*. SAN switches were replaced, which I was not informed. Of course things were not ok last night and both servers rebooted. ioscan output shows that dsfs of LUNs in EMC storage are now c17t*d* & c18t*d*. There was no vg manipulation last night, no map file saved, so I was hopeless when I finally run into this issue.
BUT, I COULD ACTIVATED ALL VGS USING vgchange -a y! And vgdisplay show pv name of all these vgs were c21t*d* & c22t*d*, which I assume no longer existed in OS. vgexport and vgimport were OK so it was easy to produce good configuration.
I don't understand it, any one does? there is powerpath on both servers.
Regards
Stephen
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-17-2013 01:33 AM - edited тАО06-17-2013 01:47 AM
тАО06-17-2013 01:33 AM - edited тАО06-17-2013 01:47 AM
Re: HW path changed, dsf changed, VGs could be activated!
That's PowerPath at work.
It remembers the device names each LUN was associated with, and allows the system to use even the "dead" DSFs, if the system has at least one good path to the LUN. This is because PowerPath takes the traffic from any and all DSFs for a particular LUN, and load balances the traffic across all the good paths for that LUN. You just saw an extreme case of that.
By the way, a map file can be generated with just the information in /etc/lvmtab, so it's no problem to export a VG whose disks are not accessible. This is a basic HP-UX LVM feature.
In fact, if you have disconnected some disks before removing/exporting their VG configuration from the OS side, exporting the now-inaccessible VG is the only clean way to get rid of the remains of their configuration.
Edit: as you've now updated the VG configuration, you probably should now run "powermt check" and clean up the old dead paths from the PowerPath configuration too.
I recall my first surprise with PowerPath was exactly like your situation too. I looked at the ioscan and LVM configuration, thought "that does not make sense", then looked at PowerPath configuration, noticed that it was tracking both the old & new DSFs, and thought "Could that be it???" Some furious RTFMing later, my conclusion was: "Huh. Looks like that's exactly it. The PowerPath sure does quite a bit of work behind HP-UX's back."
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-17-2013 05:35 AM
тАО06-17-2013 05:35 AM
Re: HW path changed, dsf changed, VGs could be activated!
Thank you for your reply!
"a map file can be generated with just the information in /etc/lvmtab", I never try this, how to get that?
Regards
Stephen
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-17-2013 02:52 PM
тАО06-17-2013 02:52 PM
Re: HW path changed, dsf changed, VGs could be activated!
Nothing special is needed for that. In fact, that is probably how the vgexport command always works.