-
1
- #1
Hi,
I've just had this problem, so I post it here for future reference.
A client has a couple of HMCs V7R3.2. After a reboot they both hang at pre-logon screen, web access also failed saying the console was not ready, and when login with ssh most commands failed with message: "connect: Connection refused. A connection to the Command Server failed."
We didn't know it, but it is a known bug in versions V7R3.1 and V7R3.2. They have a problem with log rotation, that let the log /var/hsc/log/hmclogger.log grow without limit (in this client this log grew up to 400M).
The solution is to raise a call to IBM support and obtain the pesh password. With this password you can gain root access (just for a day) to HMC and empty this log.
There's also an unsupported solution, booting the HMC from a Linux Live CD, mounting the internal disk and emptying the log, but as I've said it is not supported, and you can loose IBM support.
So now we've solved this problem, and we will have to plan the upgrade of both HMCs.
Link to this bug reference:
Hope this helps.
I've just had this problem, so I post it here for future reference.
A client has a couple of HMCs V7R3.2. After a reboot they both hang at pre-logon screen, web access also failed saying the console was not ready, and when login with ssh most commands failed with message: "connect: Connection refused. A connection to the Command Server failed."
We didn't know it, but it is a known bug in versions V7R3.1 and V7R3.2. They have a problem with log rotation, that let the log /var/hsc/log/hmclogger.log grow without limit (in this client this log grew up to 400M).
The solution is to raise a call to IBM support and obtain the pesh password. With this password you can gain root access (just for a day) to HMC and empty this log.
There's also an unsupported solution, booting the HMC from a Linux Live CD, mounting the internal disk and emptying the log, but as I've said it is not supported, and you can loose IBM support.
So now we've solved this problem, and we will have to plan the upgrade of both HMCs.
Link to this bug reference:
Hope this helps.
