A question for the empeg guys about upgrade files.

I have been playing with building upgrade images to pump /dev/hda2 and install new kernels and have had quite a lot of success... However I occasionally get the following error message and would like to know what the error means so I can work out where I am going wrong...

> upgclient -p /dev/ttyS0 -s 115200 output.upgrade
empeg-car Upgrade client.
empegupgrade: Using upgrade file 'output.upgrade' to device '/dev/ttyS0' (115200 baud)
 Checking upgrade file integrity                                        [100%]
Pumping /dev/hda2
 Finding unit                                                           [  4%]
Upgrading flash (bootloader)
 Programming flash                                                      [100%]
Upgrading flash (kernel)
 Programming flash                                                      [100%]
Upgrading flash (ramdisk)
 Programming flash                                                      [100%]
Pumping primary drive
 Selecting pump device                                                  [  0%]
Pumping /dev/hda2
 Pumping partition                                                      [ 22%]
 ERROR: 13 Pumping failure
 Upgrade aborted due to errors                                          [  0%]
Upgrade failed :-(

Upgrade failed with error 0x80040057

trying again with the same upgrade file usually results in the same error... but often not at the same point which leads me to believe it isn't something specifically wrong with my upgrade file but more to do with communications between the upgclient and the empeg...

I have just tried the same upgrade file using the Windows Upgrade Wizard and got an error:

Error "Bad pump response" occurred during stage 0x22

Any help would be much appreciated...

Cheers

Kim

(I seem to make a habit of replying to my own posts shortly after making them...)

I think I might have tracked it down to being something in the custom kernel images rather than the hard drive images...

I was using Mark Lord's patches with some small patches of my own... so it looks like I am going to have to go through the source and see what might be causing the problem... Maybe some stray printks or something...

Cheers

Kim

PS I still wouldn't mind getting some further info on what the errors mean and possible causes... it would help trying to track down the fault

(and again)

I think I have narrowed it down to the temp reading code... maybe reading it so often is interrupting th process...? Could calling init_temperature interrupt the partition pumping code...?

Cheers

Kim

It wasn't just the init_temperature... the polling of read_temperature to test against the warning temp in Mark Lord's kernels was making upgrades happen deathly slowly... and then it seemed they would just get out of sync eventually...

The fix was to turn off the max temp warning while applying an update...

Cheers

Kim

PS asking and answering your own questions is a great way to increase your post count... :)

Good catch.

Hey Tony.

Do you mean the catch about the temperature reading, or the one about answering one's own posts?

cu,
sven

I meant the catch about the temperature reading. Trying to increase your post count, too, Sven?

Trying to increase your post count, too, Sven?

Me? Never.

cu,
sven

PS: 65 posts to go for addict. I really already feal like an addict: I read (almost) all messages on this BBS.

>It wasn't just the init_temperature... the polling of read_temperature

That's very strange.. the polling is supposed to happen only once every two seconds.

Could you perhaps play with it some more to see what was going on, so I can fix it?

Thanks

-ml