This post is more than 5 years old
21 Posts
0
79840
October 1st, 2014 07:00
Issues with Out-of-Band updates
I'm currently running Dell's OME v2.0. Everything is working great including updates via OMSA. However, updates done via the iDRAC connections fail every time.
I could say I have tried everything, but I figured somebody could help me with troubleshooting everything all over again in case I missed something or did something incorrectly.
The DRAC connections live on a different subnet, but there is no firewall between these two networks. Only a router that routes traffic between the two. As I type this I came up with a few more ideas, but if you have anything else to offer that would be great.
Here is the error I get when trying to deploy the updates.
Results:
Downloading Packages.
Calling InstallFromUri method to Download packages to the iDRAC
Downloading Package: BIOS_V7P04_WN64_2.2.3.EXE onto the iDRAC
Package download has successfully started and the Job ID is JID_121890750543
The URI given to the iDRAC to download from: http://[serverip]:1278/install_packages/Packages/BIOS_V7P04_WN64_2.2.3.EXE
Checking the Status on the following Job(s): JID_121890750543
The JOBID JID_121890750543 has failed with the following message:
The network file transfer of an update package was not successful.
MessageID: RED051
Download Error occured.
Package BIOS_V7P04_WN64_2.2.3.EXE failed to download on target device. Update failed.
Software Update Task Failed.
Thanks,
b-rocol1
21 Posts
0
November 4th, 2014 08:00
Well the call with Dell didn't help, but eventually I was able to figure it out. I learned that when trying to browse out to a package (executable) for download, IIS would return a 404 error.
An example URL was http://[nameofserver]:1278/install_packages/Packages/SAS-RAID_Firmware_C3X7D_WN32_12.10.6-0001_a12.exe
I had to open up the web.config for the Essentials site and add the following snippey inside the section.
The site's web.config file is located at C:\Program Files\Dell\SysMgt\Essentials\wwwroot\Essentials\.
Once that module was in place I was able to browse to any package and it would prompt me to download it. From there, I tested updates out again on the ESXi hosts and all is well again. Hopefully this little bit of information helps someone else!
Thanks again for everyone's assistance.
DELL-Rob C
3 Apprentice
•
2.8K Posts
0
October 1st, 2014 15:00
Hi there.
Ok, so I think RED051 may indicate a network problem, did you retry? (I assume you did).
So for iDRAC updates, I'm assuming you discovered the iDRACS via WSMan yes? And they are correctly classified under the RAC group etc?
I wonder if your job queue is full? Can you run the Dell Troubleshooting tool and do the test that Lists all fhe jobs in the iDrac just to see?
Thx
Rob
delltechcenter.com/ome
b-rocol1
21 Posts
0
October 2nd, 2014 06:00
Thanks Rob. That I did try, but I'll clear out the job queue and give it another go. Looks like there are quite a few in there. We'll see what happens.
That is correct. They were discovered and inventoried using the WSMan protocol and do show up under the RAC group.
I'll post back once I clear out the job queue and give it another shot this morning.
b-rocol1
21 Posts
0
October 2nd, 2014 11:00
So I tried to rerun the updates, but received the exact same errors as before. So I took a look at the job queue and cleared it out and then rebooted the server.
Protocols Selected are:
InstanceID = JID_CLEARALL
JobStartTime = TIME_NA
JobStatus = Pending
JobUntilTime = TIME_NA
Message = NA
Name = CLEARALL
PercentComplete = 0
MessageID = NA
DCIM_LifecycleJob
InstanceID = JID_122881064600
JobStartTime = NA
JobStatus = Failed
JobUntilTime = NA
Message = The network file transfer of an update package was not successful.
Name = update:DCIM:INSTALLED#802__USC.Embedded.1:LC.Embedded.1
PercentComplete = NA
MessageID = RED051
DCIM_LifecycleJob
InstanceID = RID_122881078124
JobStartTime = TIME_NA
JobStatus = Pending Reboot
JobUntilTime = TIME_NA
Message = New
Name = Reboot1
PercentComplete = NA
MessageID = JCP000
After deleting all jobs via the Dell Troubleshooting Tool, I noticed this job remains in there. I assume this isn't causing issues with the updates correct?
I can ping the OME server from each of the drac connections, so that's a good sign.
Anything else I could be missing. Again, thanks for the help!
DELL-Rob C
3 Apprentice
•
2.8K Posts
0
October 3rd, 2014 08:00
yeah, I think your job queue is ok, now. Sometimes folks will do a racreset on the off chance that helps.
We may need you to open a ticket so the support guys can take a closer look at that iDRAC and see if they can see whats up (800-945-3355).
Rob
b-rocol1
21 Posts
0
October 6th, 2014 07:00
Sounds good. Will do.
On a separate note, is it possible for OME to send out email notifcations/alerts when new firmware, BIOS, drivers, applications, are available for a server?
DELL-Rob C
3 Apprentice
•
2.8K Posts
0
October 6th, 2014 08:00
So OME does not have this feature. But I noticed on support.dell.com it seems to have something kind of like this. I wonder if you can try it and let me know.
Go to support.dell.com
Then go to PowerEdge and choose your model of server
Then pick a category like BIOS
I see a link for Drive Update Notification (via email).
Does that help? Not integrated into OME I realize, but maybe useful.
Thx!
Rob
nadogmoney
2 Intern
•
184 Posts
0
October 10th, 2014 09:00
I don't use OME but does it do JID_CLEARALL before updates? What about forcing the LC State to enabled? I just added both these additional workarounds to our online DUP method of updating and they seem to take care of a lot of issues.
+ start rant
"Drive Update Notification" via email doesn't work for at least the past 4 years. We have complained for years to our Dell sales guys. We used to get monthly update notifications from our sales guys called a PDE but I don't think we get that anymore. Driver release management and notification from Dell has a lot to be desired. "Block" release components can get posted to the site weeks apart and I have never seen any place that states what components are tested and included in a "block" release.
+ end rant.
b-rocol1
21 Posts
0
October 10th, 2014 14:00
Thanks Rob. I may go that route although I'm not too fond of the idea. I know Repository Manager sort of gives you the feature too, but I'd rather have it implemented right within OME as it would allow everything to be configurable and manageable through a single of glass persay.
@Nadogmoney
What do you have to do to force the LC state to enabled? Haven't been able to figure that one out yet.
DELL-Rob C
3 Apprentice
•
2.8K Posts
0
October 10th, 2014 14:00
Thanks and I'll fwd the rant along...the feedback is appreciated.
Not sure about the enabled. But as far as CLEAR, I think we have to be careful before doing things like that unexpectedly against an iDRAC. I'm sure it was discussed but there might have been customer concerns as to that happening at the wrong time. But will pass this along as well.
Thanks,
Rob
nadogmoney
2 Intern
•
184 Posts
0
October 13th, 2014 07:00
@b-rocol1
In 12+ Gen:
racadm set LifecycleController.LCAttributes.LifecycleControllerState Enabled
I am asking the same Q for 11 gen:
http://en.community.dell.com/techcenter/systems-management/f/4469/t/19602755
@rob c
We have lots of workarounds to get online configs and DUPs installed. CLEARALL is just the latest edition. I doubt it is possible but I would love to see all the workarounds needed under the covers of OME to get configs and FW installed.
DELL-Pupul M
2 Intern
•
1K Posts
0
October 14th, 2014 00:00
Hi,
This link should help you:
http://en.community.dell.com/techcenter/systems-management/f/4494/p/19518120/20420677#20420677
It is not possible to do it for 11G server using racadm. You will have to do it through the Ctrl E option.
nadogmoney
2 Intern
•
184 Posts
0
October 14th, 2014 15:00
@ Pupul
Are you taking about CSIOR or setting the LC state to enabled?
Maybe I missed something in that link but I see someone incorrectly saying CSIOR can only be enabled with Ctrl-E in 11th gen. It can definitely also be enabled with WSMAN.
DELL-Pupul M
2 Intern
•
1K Posts
0
October 15th, 2014 10:00
Well yes, i think the post wanted to confirm that racadm cannot be used.
I am talking about CSIOR. When the server reboots, the system inventory needs to be collected so that OME can read the correct information from iDRAC.
nadogmoney
2 Intern
•
184 Posts
0
October 16th, 2014 08:00
I agree enabling CSIOR can not be done in 11g with racadm but I disagree that Ctrl E is the only other option. Setting CSIOR can definitely be set by via WSMAN. I do it currently.
1. Set CSIOR:
winrm i SetAttribute http://schemas.dmtf.org/wbem/wscim/1/cim-schema/2/root/dcim/DCIM_LCService?SystemCreationClassName=DCIM_ComputerSystem+CreationClassName=DCIM_LCService+SystemName=DCIM:ComputerSystem+Name=DCIM:LCService -file:$dir\CSIOREnable.xml -r:https://$IDRAC:443/wsman -u:$USER -p:$PASS -auth:basic -encoding:utf-8 -SkipCNCheck -SkipCACheck -skiprevocationcheck -format:xml
File CSIOREnable.xml:
Collect System Inventory on Restart
Enabled
2. Create a Job:
winrm i CreateConfigJob http://schemas.dmtf.org/wbem/wscim/1/cim-schema/2/root/dcim/DCIM_LCService?SystemCreationClassName=DCIM_ComputerSystem+CreationClassName=DCIM_LCService+SystemName=DCIM:ComputerSystem+Name=DCIM:LCService -file:$dir\CreateConfigJob.xml -r:https://$IDRAC:443/wsman -u:$USER -p:$PASS -auth:basic -encoding:utf-8 -SkipCNCheck -SkipCACheck -skiprevocationcheck -format:xml
File CreateConfigJob.xml:
00000000002200.000000:000
false
Easy, right ?!?..... NOT!
At least it is not via console and Ctrl E. Now WSMAN just needs to learn how to do power drains!