Solved!
Go to SolutionOneFS 8.0.0.5 isi upgrade assess problems
I have a cluster running 8.0.0.4 and I am planning on upgrading it to 8.0.0.5 due to a tcp security vulnerability.
I ran the ' isi upgrade cluster assess /ifs/data/OneFS_v8.0.0.5_Install.tar.gz'
but I never received any output back (waited 30 minutes).
I then ran the 'view' option and then the 'isi_upgrade_logs -a' option.
# isi upgrade cluster view
Upgrade Status:
Cluster Upgrade State: committed
Current Upgrade Activity: Upgrade
Upgrade Install Image: /ifs/data/OneFS_v8.0.0.5_Install.tar.gz
Upgrade Start Time: 2017-06-30T09:44:05
Current OS Version: 8.0.0.4_build(53)style(5)
Upgrade OS Version: N/A
Nodes Progress:
Total Cluster Nodes: 7
Nodes On Older OS: 7
Nodes Upgraded: 0
Nodes Transitioning/Down: 0
If there are any errors please run "isi_upgrade_logs" to gather more information.
# isi_upgrade_logs -a
For Assessment with Upgrade ID [2017-06-30T13:44:03.000Z]
Assessment was successful
Should I assume that there were NO issues found? or did the assessment fail to successfully run?
addisdaddy20
65 Posts
5491
1
Posted July 7th, 2017 13:00
If you like my customer are disappointed in the lack of serial output for upgrades, I would suggest getting with your account team to file a Feature request. I know my account team did that for our customer because they basically had the same type of issue. Lack of output to understand where the upgrade is at or doing including success or failures. The fact that you have to query to find either can waste lots of time if it failed right after I checked it was running it could potentially add an extra 1/2 hour to maintenance window on top of whatever amount of time it takes to fix the issue.
Naturally you could script something to dump the commands output on screen but its irritating when this kind of feature was there and now its not. FRE kind of feel like a black hole but it can never happen without one. Currently that is the best method to let product management understand that customers want that functionality back so your not guessing what an upgrade of any kind is doing.
sjones5 is correct, create an SR for upgrade. Remote Proactive will give a better assessment & process. Check the Isilon Uptime Info Hub out also, it has a section on upgrading.
...
Upgrading
- OneFS Tip - Reboot Before Upgrading
- Upgrading from OneFS 6.5: Ensuring a Smooth Upgrade Process
- Upgrading to OneFS Target Code
- Limiting downtime for SMB workflows during OneFS Upgrades
- Improve Cluster Reliability by Upgrading Firmware
- Update Your Firmware: Keep the Gnomes on Track
- Planning OneFS Upgrades with Global Namespace Acceleration (GNA) Nodes
The document tells you to run the pre-check but doesn't tell you that you need to perform a manual step to check the results. At the very least, the step to manually check via cli "isi_upgrade_logs -a" should be included in the pre-check documentation. All the documentation says is "the utility may take several minutes to run. If the utility returns errors, resolve the errors before continuing with the upgrade."
The problem is, you run the command, it returns automatically, and then nothing else is returned. There is no mention to manually check the log files.
The document I am referring to is "docu69989_OneFS-8.x-Upgrade-Planning-and-Process-Guide.pdf" n page 36.
Jim
dfelt1
34 Posts
5493
0
Posted July 7th, 2017 12:00
Hello Jim Kunysz,
There were changes in how OneFS 7.x and *.x does their assessment for upgrading. The assessment is still run during the pre-upgrade checks, the difference is how the output is accessed. For OneFS 8.x from the CLI use "isi_upgrade_logs -a" to get the information. Additional information can be gathered but requires digging through the logs.
Is there anything specific you are looking for? I can always inquire with engineering on how to find specific info now for the 8.X upgrades.
Thank you,
David