Defective Memory - Primary Xen Server
-
So my primary xen server has defective ram in it, yesterday the Dom0 of this host was hung, and the plan was to reboot it this morning.
Well didn't get the chance, the host crashed last night.
Good thing I have a tiny box to migrate my VM's over too.
Rather than trying to rebuilt the 3 critical VM's from scratch, import my backup from Sunday.
Yah I win!!
-
How did you know it was RAM? Assuming you have more than one stick in there, couldn't you just pull the seemingly likely single bad chip?
-
Well it's either RAM (4 short fast BIOS Beeps) or the board is bad.
Which would suck.
-
And it's not making it past POST. I have testing to do.
-
@DustinB3403 said:
And it's not making it past POST. I have testing to do.
oh.. yeah.. not cool - good luck.
-
But the critical point is that it's stupidly simple to recover from this.. Stupidly simple.
VM Says: "Oh you have another Xen Server, yeah I can run over there in the mean time. "
-
XenServer for the win.
-
So this is what I'm not present with, the server finally gets past post, but won't get past this.
Anyone have any pointers?
-
@DustinB3403 said:
So this is what I'm not present with, the server finally gets past post, but won't get past this.
Anyone have any pointers?
That doesn't look good. It looks like it's not seeing md0. You might have to boot from a USB and do some troubleshooting.
-
Yeah (fortunately) this system was hobbled together so it's single disks.
I'm going to attempt to mount them to a laptop running *nix and see what I can pull off.
-
@DustinB3403 said:
Yeah (fortunately) this system was hobbled together so it's single disks.
I'm going to attempt to mount them to a laptop running *nix and see what I can pull off.
Assuming a drive failure, try running SpinRite on it.
-
I'm thinking it was a memory issue, so the data on them might be OK.
-
@DustinB3403 said:
Yeah (fortunately) this system was hobbled together so it's single disks.
I'm going to attempt to mount them to a laptop running *nix and see what I can pull off.
Wait it wasn't raid?
-
@DustinB3403 said:
I'm thinking it was a memory issue, so the data on them might be OK.
How did you get past the memory issue, yet still have the error if it was a memory issue?
-
@johnhooks no...
It's a long story
-
-
-
This was the production server that was budgeted together... fortunately I have recent backups of my VM's so now it's just a matter of rebuilding it.
-
This was also before our topic on MDAD RAID 10.
-
@DustinB3403 said:
This was also before our topic on MDAD RAID 10.
Are you going to rebuild it MDAD RAID 10?
