Openscape 4000 Assistant 1

[Gandalf1454]

Our facility lost power twice yesterday and has affected our phone system although connected via UPS. On both of our Openscape 4000 ecoservers the fail light is red and I need help determining what is down and what to do. I'm familiar with some siemens cmd line, appbuilder and openscape assistant however I've never dealt with this before. Below is a snippet of the screens on both ecoservers from the platform portal. We monitor hardware via PRTG and 10.1.25.2 is currently down and is named OS 4000 assistant. For reference, 10.1.25.1 is also listed as assistant in the LAN overview page in platform portal.

Any and all help is appreciated.

 

[Moriendi]

Everything looks good on your screenshot, and with an uptime of 78d, we know it didn't reload with your power problem.
The red LED just means you have a major alarm. From comwin sta-hista:mirror; will tell you which alarms are on.
If you can browse to the Assistant, it's OK, but if you can't ping it even internally from the platform, I'd probably just tell it to reinstall from the Portal.

 

[sbcsu]

The 'fail' light is not a fail light - it's just an indication that there is an urgent alarm on the system.
DIS-SIGNL:SYSTEM; will tell you more
If you're happy and don't want to check then delete the alarms.
DEL-GRA:A1,ALL;
DEL-GRA:BP,ALL;

To check the Assistant you should log on via the Atlantic LAN - the Assistant is at 192.0.2.5

 

[Gandalf1454]

Hi Moriendi, thanks for your response. I ran the cmd you mentioned and below is the output. I cannot ping 10.1.25.2, can you tell me where I need to reinstall within the portal?

Hi sbcsu, thanks for responding as well. I ran your cmd and below is the output, please advise.

 

[Moriendi]

The power supply problem is worth looking at. If you have the AP3700 shelves with the LUNA PSUs, each one will have a green LED if it's working. Looks for one that's either powered off by accident, or powered on with no LED. Something powered on with no LED you could try an off/on, and if that doesn't help, take the front plate off (six torx screws) and just slide it out, it will need replacing.

The Assistant should be kept running by the cluster management. It should be pingable both internally and externally on the 10. IP you had, and also internally on the 192.0.2.5 address sbcsu gave you, so if you know the root password you could ssh to the portal IP and try and ping it from there.

"crm_mon -1fA" from ssh will tell you if that resource is up and running.
Resource Group: gr4k_Assistant
rsc4k_fs_Assistant (ocf::heartbeat:Filesystem): Started cora-cca
rsc4k_Assistant (ocf::soco2:LXC): Started cora-cca
+ rsc4k_Assistant-RASTATE : ASSISTANT_RUNNING

It could be that maybe it has tried to reinstall itself and got stuck, and it can't install. You can try and force a reinstallation from Portal, Maintenance.

If it's not pingable at all, and you make a "df -kh" from ssh, what is the free space on /opt/Assistant/share? And if it's nearly full, what are the contents of that directory ? You might find the free space is changing if it's trying to reinstall so check twice 5 minutes apart.