Reboot paranoia 1

[Mag0007]

In the last 4 weeks, I basically went thru hell! How do you guys verify if your system will comeback up again well? Here is the checklist I have been following:

Bootlist is set to hdisk0 hdisk1
errpt shows no 'P' errors
Perform a clean shutdown (shutdown -Fr)

These are the errors I recently faced:
0516/0517/0518 (rootvg filesystems are corupted, eventhough I did a clean shutdown)
The boot image was not found
The horrid looping of the SMS menu.

What have you guys faced during reboots? Just curious, wanted to hear some war stories.

 

[mrn]

Hi Mag

Quite interested in this post, could you give us a bit more info.

What version of AIX?

Full error codes

How did you fix?

Mike

"A foolproof method for sculpting an elephant: first, get a huge block of marble, then you chip away everything that doesn't look like an elephant."

http://www.airport-parking-site.co.uk/

 

[RodKnowlton]

Mag,

I suppose you could get a little more assurance by doing new bosboots on the drives before rebooting. And of course a mksysb.

No war stories here (knock wood). The closest I've come to a problem rebooting was a device causing it to hang during configuration, but all I had to do was disconnect the device.



Rod Knowlton

IBM Certified Advanced Technical Expert pSeries and AIX 5L
CompTIA Linux+
CompTIA Security+

 

[p5wizard]

Or a service mode boot that goes on and on while scanning all the SAN disks in search of disks with a valid boot record.

SSA device driver at a certain level in AIX433 with certain types of SSA adapter/disk combination also gave boot hangs at LEDcode 80c if I'm not mistaken...

But yes, if you perform a bosboot and it doesn't complain about problems which would hinder a safe reboot of the machine, you should be OK. Also check that the disks in the boot list contain the hd5 mirrors!


HTH,

p5wizard

 

[Mag0007]

Hi Mrn:

AIX version: 4.3.3, 5.1, 5.2, 5.3 :-(

I actually forget some of the error codes, I was too busy and working frantically to fix the problem. Here are some fixes, by order which I used the most:

#1) Get the AIX CD #1 , and boot off that: With this I have cleaned by the rootvg filesystems (log check, fsck, bosboot, bootlist). Saved me many many times.

#2) Updated the firmware, I know its a little wierd, but yes updating to the latest firmware has fixed it before

#3) Mksysb, NEVER worked for me yet. Thats probally why I hate using tapes to do restores. Next time I want to do this via network (NIM), i think that would be cool.


p5wizard:
I am not sure what you mean by the hd5 mirror, I have heard something similar before but didn't know what they were talking about (naturally I was thinking of drinking beer, and smoking)

Some questions I have is:
Is there a way to verify if there is an active bootimage on the hdisk? maybe a bootinfo with a flag.

TIA

 

[p5wizard]

if you have bootlist hdisk0 and hdisk1, just make sure that lslv -m hd5 shows these disks where the partition mirrors are:

# lslv -m hd5
hd5:N/A
LP PP1 PV1 PP2 PV2 PP3 PV3
0001 0001 hdisk0 0001 hdisk2

As to hd5 being valid: build a new one and make sure there's no errors

bosboot -a -d /dev/hdiskX -l /dev/hd5

Substitute X for 0 or 1 or whatever your hd5 primary mirror is on. Make sure rootvg's mirrors are synchronized

# lsvg -l rootvg|head -2; lsvg -l rootvg|grep hd5
rootvg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT POINT
hd5 boot 1 2 2 closed/syncd N/A

if not syncd - use
syncvg -v rootvg


HTH,

p5wizard