| Author |
Broken Alpha 2100 |
esiciensky
Member
Posts: 7
Joined: 15.06.18 |
| Posted on June 15 2018 09:30 |
|
|
I had to replace a CPU module in my Alpha 2100 - we replaced with the same CPU module version that we had...a B2020
We have not been able to get the machine to boot after the replacement, and the message is "Failed to Open dka0.0.0.1.0" which is the boot dev.
Is there something I needed to do after replacing the CPU module that I am not doing ? Any help would be appreciated, this is a really old machine, but we need to get it running to pull off some needed historical data.
Thanks |
|
|
| Author |
RE: Broken Alpha 2100 |
abrsvc
Member
Posts: 108
Joined: 12.03.10 |
| Posted on June 16 2018 02:19 |
|
|
At first glance, I would suspect that the cables are not connected the same. At the console, check for disks using the SHOW DEV D command. Verify that you are seeing disks for one, then verify that the disks are the same as the old config. In other words, did you connect the cables in the same manner? Perhaps DKA0 is now DKB0?
Dan |
|
|
| Author |
RE: Broken Alpha 2100 |
esiciensky
Member
Posts: 7
Joined: 15.06.18 |
| Posted on June 16 2018 03:18 |
|
|
I don't think we pulled any cabling..just pulled out the cpu card and put the replacement using the same slot ( it is a single cpu system)...when I do a show device, the boot disk is displayed with the same dka0 name and matches the boot_dev info. is there anything else I can try...and thanks for your reply
Emily
|
|
|
| Author |
RE: Broken Alpha 2100 |
abrsvc
Member
Posts: 108
Joined: 12.03.10 |
| Posted on June 16 2018 03:37 |
|
|
If the drive is seen as the same device, then the on-board controller may have a problem. I believe that the console has a disk test that does not alter/touch the user data. I would run this test. The drive is "seen", but not available per the message. If the device description matches what existed before, then this means that there is no communication between the device and the cpu. The device can be seen without any access to the actual disk. This means that the controller is acknowledging the existence only.
Try the disk test and post the results here.
Dan |
|
|
| Author |
RE: Broken Alpha 2100 |
esiciensky
Member
Posts: 7
Joined: 15.06.18 |
| Posted on June 16 2018 04:51 |
|
|
|
Thanks Dan..I will run the risk flags and post them.. |
|
|
| Author |
RE: Broken Alpha 2100 |
esiciensky
Member
Posts: 7
Joined: 15.06.18 |
| Posted on June 16 2018 15:59 |
|
|
|
Something came to me...the replacement cpu...does it have dip switches or connectors tell the board where the boot device is..we did not check that..we just made sure it was a b2020...should I post this under the hardware forum? |
|
|
| Author |
RE: Here is the test disk function results |
esiciensky
Member
Posts: 7
Joined: 15.06.18 |
| Posted on June 18 2018 03:28 |
|
|
Testing the Disks (read only)
failed to send Read to dka0.0.0.1.0
*** Hard Error - Error #5 -
Diagnostic Name ID Device Pass Test hard/Soft 1-Jan-2066
exer_kid 00000180 0 0 1 0 12:00:01
Error in read of 0 bytes at location 00002A00 from device dka0.0.0.1.0
*** End of Error ***
Then is tries to boot
(boot dka0.0.0.1.0 -flags 0)
failed to open dka0.0.0.1.0 |
|
|
| Author |
RE: Broken Alpha 2100 |
abrsvc
Member
Posts: 108
Joined: 12.03.10 |
| Posted on June 18 2018 11:26 |
|
|
Looks like the controller on the CPU board has failed. Note that the device is detected, but a read did not make it to the drive. I think that there are 2 scsi channels on that board. Try the other one. You will ned to boot dkB0 though.
Dan |
|
|
| Author |
RE: Another running Alpha |
esiciensky
Member
Posts: 7
Joined: 15.06.18 |
| Posted on June 19 2018 08:25 |
|
|
Dan - thanks so much for your help - when you mentioned the controller - it occurred to me I should try just reseating all the ribbon connectors on the I/O board..to my great relief - after doing this, the Alpha booted up...as an old "DECie" I should have thought of that first..
thanks again- I was glad at least to have found this interesting forum
Emily
Edited by esiciensky on June 19 2018 08:26 |
|
|
| Author |
RE: Broken Alpha 2100 |
abrsvc
Member
Posts: 108
Joined: 12.03.10 |
| Posted on June 20 2018 01:51 |
|
|
A partially connected cable can easily explain your symptoms. Enough of the pins were present to detect the device, but not enough to actually perform work. This is more common than you would think with those types of connectors.
Glad to know that this was resolved.
Dan |
|
|
