I have a HP-UX B.11.11.
I had one disk and added one new.
When trying to configure the second disk Not Using the Logical Volume Manager(from SAM) I have this error:
Pid X killed due to text modification or page I/O error
I tryed to add another partion on the first disk, but I ended with to same error. I managed to add another swap partition with success on the first disk.
I runed xstm on both disks to check them for harware failures but it seems that they are OK.
I saw also a strange message in system log:
Synchronous Page I/O error occurred while paging to/from disk
Can anybody help me with this issue?
Thank you in advance!
Stefan
To begin to help, we would need to know a bit more:
Architecture (server -WS )? PA-RISC?
Not using LVM means what? (Not using SAM isnt the same thing...)
What were the commands you passed (and why maybe...).
this is a HP-UX workstation:
HP-UX hostxxx B.11.11 U 9000/785 with a PA-RISC CPU.When you try to configure a disk in SAM you can choose with LVM or without LVM (Not Using the Logical Volume Manager).
I wanted to use the VxFS that is why I did not choose the LVM.
I did not passed any commands because I did it from the gui of the SAM.
VxFS or VxVM ?
All our servers use LVM but the filesystems are VxFS (OnlineJFS)...
In other words 2 totally diferent products...but that not the todays topic I agree...
2 things come to my mind:
text modification -> Any NFS around?
or
I/O error -> That may not be good news...
I would test the (yet?) primary swap:
fsck -F vxfs -y -o full /dev/vg00/lvol2
But I dont know what it would look like when not in LVM... (the case above)
You could also try:
I use NFS, but I still got the error after a disable of the NFS client and reboot.
The command for checking swap returned:
bash-2.04# fsck -F vxfs -y -o full /dev/vg00/lvol2
vxfs fsck: not a valid vxfs file system
invalid super-block
search for auxiliary super-block? (ynq)y
alternate super-block not found
vxfs fsck: cannot initialize aggregate
file system check failure, aborting ...
bash-2.04#
I tried also the "dd" ideea -> the HDD has 73 GB -> but i runned about 4-5 hours, then I breaked the process. Durnig that I examined the syslog file. It seems that a system monitor (disk_em) wrote a log it a error. I was impossible to read some date from disk. The event code is 100337. On the HP site they say that I have reformat, replace .
Thanks for the help until now, do you have any ideeas next ?
Regards,
Stefan
OK what does bdf, vgdisplay vg00 gives? whats in /etc/fstab /etc/mnttab
Your case is difficult to figure yout because dont know much about the system, what it does, what its for etc... and how it is configured.
From what you said previously, I would go and remove the new added disk, if any nfs - unmount it. And check the configuration and state of the box.
At this point, that means minimalist with just vg00 as boot disk - does it still gives any error? you have the log files + stm (xstm) to check.
If thats OK what do you want to do with the added disk?
Add to vg00 or create a new volume group.
If not OK what kind of disk is the new one? (internal -external?)
Keep in mind there are numerous SCSI standards (HVD-LVD/SE/E../FW...)
which can lead to severe ssci issues...
What I have in mind is if same or bigger disk doesnt mind the lost space since it is to find whats going on for now... copy with dd the first to the second and boot from it and see if this solves your issue, if so you know now you have a bad disk...
About NFS...
Are you using multiple NICs ?
I have seen cases where communication gets messed up when going out through one and entering through the other...