So, I’ve set up OSD Bare Metal several times but I’ve never seen this problem…
PXE Boots fine, the install is moving along… copies the MDT Bundle, OS Resources, but them bombs on the WIM file. I’ve confirmed all of the files (including WIM) are on the bigfix server and OSD / Relay system. the SHA1 values match perfectly.
The bigfix server, bare metal server, and target are all on the same vSphere host.
BigFix : 9.5.1.9
Bare Metal : 7.1.120.28074 on Windows 2008R2
Here is a portion of the logs, showing the OS Resource copy (successful), then the WIM (fail):
[2016/05/12 23:34:16] W <INF> Remote copy for [OSResources] started. Name [1EF7D2CCADA32990C5BBD52476693B2CB1F5B0EA.zip], size [411592708], remote [tem://bigfix.domain.com:52311/Uploads/1ef7d2ccada32990c5bbd52476693b2cb1f5b0ea/1EF7D2CCADA32990C5BBD52476693B2CB1F5B0EA.zip.BFOSD/1EF7D2CCADA32990C5BBD52476693B2CB1F5B0EA|411592708|32AD8EBEE0C981C1461135712B1C7D93C42D4A562A8468A9C8EECC0D96A058ED/1EF7D2CCADA32990C5BBD52476693B2CB1F5B0EA.zip], local [local://root/C$/Deploy_work/1EF7D2CCADA32990C5BBD52476693B2CB1F5B0EA.zip]
[2016/05/12 23:34:24] W <INF> Remote file copy completed.
[2016/05/12 23:34:24] W <INF> Remote copy for [Wim File] started. Name [Win2012R2x64SP0_1462889573.wim], size [4769873716], remote [tem://bigfix.domain.com:52311/Uploads/497bb0c9bfb169a0889dfbb52864d4af5e61ef0d/Win2012R2x64SP0_1462889573.wim.BFOSD/497BB0C9BFB169A0889DFBB52864D4AF5E61EF0D|4769873716|781F57B3642F4285A09A551440CB2A86FADD49EF3418A24BE8C6222D9BC8E67C/Win2012R2x64SP0_1462889573.wim], local [local://root/C$/Deploy_work/Win2012R2x64SP0_1462889573.wim]
[2016/05/12 23:34:24] W <ERR> Error raised by CopyFile in utils.rbc, line 2966 [TMF:817]
[2016/05/12 23:34:24] W <ERR> No such file or directory (781F57B3642F4285A09A551440CB2A86FADD49EF3418A24BE8C6222D9BC8E67C)
Hi, this is a problem with download of large files with version 9.5.0 and 9.5.1 of the relay component.
It will be fixed shortly with 9.5.2. In the meantime, you can downgrade the relay on your Bare Metal server computers to 9.2.x.
Thanks
Thanks for the quick reply. I know OSD is currently plagued with this issue and the SQLEXPRESS download problem… Been that way for a couple of weeks now, crippling OSD unless one performs manual workarounds… Any idea when both of these will be fixed? Are there APAR’s tracking these?
Yes.
For the MSSQL unavailability issue, we have APAR IV84425, and the fix will be released via a OSD site refresh at the very beginning of next week.
For the large file download issue, APAR is IV84436, that will released be with platform 9.5 patch 2 (9.5.2). An official date for this cannot be communicated as of now, but it will be released soon.
Thanks
1 Like
I rolled back to 9.2.7.53 and I am still having the same issues.
‘Remote file copy failed. Unable to download [Wim File]’ reported at endpoint.
Below is twhat appears to be the pertitnent information from the deployment log. Any additional help you could offer would be greatly appreciated.
[2016/05/16 12:35:25] W Unicast Transfer for 3 files starting...
[2016/05/16 12:35:25] W Remote copy for [MDTBundle] started. Name [720D256A637D17EB06D9DA8B11CC43484CECD779.zip], size [118892290], remote [tem://usdca-tem-p50.biomet.local:52311/Uploads/720D256A637D17EB06D9DA8B11CC43484CECD779/720D256A637D17EB06D9DA8B11CC43484CECD779.zip.BFOSD/720D256A637D17EB06D9DA8B11CC43484CECD779|118892290|6B785214A7A6D392CD7FFD64FA915E733819BC2FDFBC08A81FDE88E302028274/720D256A637D17EB06D9DA8B11CC43484CECD779.zip], local [local://root/C$/Deploy_work/720D256A637D17EB06D9DA8B11CC43484CECD779.zip]
[2016/05/16 12:35:28] W Remote file copy completed.
[2016/05/16 12:35:28] W Remote copy for [OSResources] started. Name [EA645EACEA00744D0742814B540A948FECF7CE14.zip], size [258398447], remote [tem://usdca-tem-p50.biomet.local:52311/Uploads/EA645EACEA00744D0742814B540A948FECF7CE14/EA645EACEA00744D0742814B540A948FECF7CE14.zip.BFOSD/EA645EACEA00744D0742814B540A948FECF7CE14|258398447|222405B6A6600FB133C45D954450200A45D484336C8E3F5DF256F00D61CD3108/EA645EACEA00744D0742814B540A948FECF7CE14.zip], local [local://root/C$/Deploy_work/EA645EACEA00744D0742814B540A948FECF7CE14.zip]
[2016/05/16 12:35:31] W Remote file copy completed.
[2016/05/16 12:35:31] W Remote copy for [Wim File] started. Name [Win7x64-Combined.WIM], size [11171753096], remote [tem://usdca-tem-p50.biomet.local:52311/Uploads/c12dbc1ed7152d2144fa20975477af58d121a389/Win7x64-Combined.WIM.BFOSD/C12DBC1ED7152D2144FA20975477AF58D121A389|11171753096|0CFBBC6FC034264A328BDED6D1DD49774076E8F0CDF73BB261B6CAE717BDA73E/Win7x64-Combined.WIM], local [local://root/C$/Deploy_work/Win7x64-Combined.WIM]
[2016/05/16 13:35:34] W Error raised by CopyFile in utils.rbc, line 2966 [TMF:817]
[2016/05/16 13:35:34] W No such file or directory (0CFBBC6FC034264A328BDED6D1DD49774076E8F0CDF73BB261B6CAE717BDA73E)
[2016/05/16 13:35:34] W (called from getRemoteFile (actions_builtin.rbc:3996))
[2016/05/16 13:35:34] W (called from NewPxeTEMFlowAction_DoIt (actions_builtin.rbc:4558))
[2016/05/16 13:35:34] W (called from HostActivityElement_RunAction (hostactivity.rbc:1051))
[2016/05/16 13:35:34] W (called from ActivityPlan_RunAction (activity_plan.rbc:64))
[2016/05/16 13:35:34] W (called from run (sequencer.rbc:3858))
[2016/05/16 13:35:34] W (called from RAD_CheckForActivity (load.rbc:22742))
[2016/05/16 13:35:34] W (called from AgentDispatch (rbagent.rbc:4387))
[2016/05/16 13:35:34] W (called from --toplevel-- (rbagent.rbc:4814))
[2016/05/16 13:35:34] W Remote file copy completed.
[2016/05/16 13:35:34] W An Exception occurred.
[2016/05/16 13:35:34] W Error raised by getRemoteFile in actions_builtin.rbc, line 4008 [:0]
[2016/05/16 13:35:34] W Unknown error (Remote file copy failure. Unable to download [Wim File] URL [tem://usdca-tem-p50.biomet.local:52311/Uploads/c12dbc1ed7152d2144fa20975477af58d121a389/Win7x64-Combined.WIM.BFOSD/C12DBC1ED7152D2144FA20975477AF58D121A389|11171753096|0CFBBC6FC034264A328BDED6D1DD49774076E8F0CDF73BB261B6CAE717BDA73E/Win7x64-Combined.WIM])
[2016/05/16 13:35:34] W (called from NewPxeTEMFlowAction_DoIt (actions_builtin.rbc:4558))
[2016/05/16 13:35:34] W (called from HostActivityElement_RunAction (hostactivity.rbc:1051))
[2016/05/16 13:35:34] W (called from ActivityPlan_RunAction (activity_plan.rbc:64))
[2016/05/16 13:35:34] W (called from run (sequencer.rbc:3858))
[2016/05/16 13:35:34] W (called from RAD_CheckForActivity (load.rbc:22742))
[2016/05/16 13:35:34] W (called from AgentDispatch (rbagent.rbc:4387))
[2016/05/16 13:35:34] W (called from --toplevel-- (rbagent.rbc:4814))
[2016/05/16 13:35:34] W Task raised an exception
[2016/05/16 13:35:34] W Unknown error [:0] in file actions_builtin.rbc, line 4008
[2016/05/16 13:35:34] W Remote file copy failure. Unable to download [Wim File] URL [tem://usdca-tem-p50.biomet.local:52311/Uploads/c12dbc1ed7152d2144fa20975477af58d121a389/Win7x64-Combined.WIM.BFOSD/C12DBC1ED7152D2144FA20975477AF58D121A389|11171753096|0CFBBC6FC034264A328BDED6D1DD49774076E8F0CDF73BB261B6CAE717BDA73E/Win7x64-Combined.WIM]
[2016/05/16 13:35:34] W called from NewPxeTEMFlowAction_DoIt (actions_builtin.rbc:4558)
[2016/05/16 13:35:34] W called from HostActivityElement_RunAction (hostactivity.rbc:1051)
[2016/05/16 13:35:34] W called from ActivityPlan_RunAction (activity_plan.rbc:64)
[2016/05/16 13:35:34] W called from run (sequencer.rbc:3858)
[2016/05/16 13:35:34] W called from RAD_CheckForActivity (load.rbc:22742)
[2016/05/16 13:35:34] W called from AgentDispatch (rbagent.rbc:4387)
[2016/05/16 13:35:34] W called from --toplevel-- (rbagent.rbc:4814)
[2016/05/16 13:35:35] W Action Preparing deployment binaries failed.
I am now seeing this at two different customer’s sites. I am going to try and rollback our onsite relays back to 9.2.7 and see how we go. Just when I thought OSD was working solid 
What I had to do to fix it:
- Uninstall the bare metal server
- Run the OSD Cleanup action
- Uninstall the Relay
- Reboot
- Install 9.2.7 Relay
- Install OSD
- Link Profiles
- Full Sync
Whether or not I HAD to do all of those who knows… But I do know that just downgrading the relay in place did not work for me…
FWIW…
1 Like
All of your steps appear to be fairly self explanatory, except for #7. Can you give me a bit more information on what you mean by ‘Link Profiles’ ?
Thanks,
Tim
When you create a bare metal profile it has to be added (linked) to the bare metal server using the server management dashboard.
Systems Lifecycle -> OS Deployment and Bare Metal -> Manage Bare Metal Servers -> Server Management
then select the bare metal server and click the “Add Profie” button in the profiles section below. This will show the bare metal profiles you’ve defined in the Image Library Dashboard. Select the ones you want on the selected server and save. That should action a profile sync to the bare metal server.
Hi all, platform patch 9.5.2 is out.
Please also make sure that your relay cache is large enough (50GB)
Chad, first off I wanted to say, you’re awesome. Thank you for taking the time to try to help me through this issue. Unfortunately, you have been more responsive to my queries than IBM support has.
The really sad part is that even after I followed your steps yesterday afternoon/evening, the issue persists. I started adding the profiles to the OSD Server yesterday afternoon around 6pm. A single WIM with two profiles. This morning at 9am the action is still listed as “pending downloads”. Apparently, I am still unable to push large files around through IEM.
Any direction that could be offered from any source would be greatly appreciated right now.
Thanks,
Tim
I should have refreshed the post before typing me last response. I will absolutely be installing 9.5.2 today.
9.5.2 is working like a champ! Thanks!
Chad, I wish I could report the same. 9.5.2 has somehow made my IEM environment worse. Assets do not check in regularly, server resources are being pegged, fixlets fail to deploy with an “Unexpected HTTP response: 404 Not Found” as the most common error, and the OSD issues are still present. Super fun week, so far. Our PMRs are sev 1. I’m sure IBM support is working hard, but we aren’t getting anywhere and I’m really starting to feel the fire from the business.
From your earlier post:
Is the file actually there? You can open a browser, and connect to http://usdca-tem-p50.biomet.local:52311/Uploads/c12dbc1ed7152d2144fa20975477af58d121a389/Win7x64-Combined.WIM.BFOSD/C12DBC1ED7152D2144FA20975477AF58D121A389 from a client that has access to the root server (or even open the browser on the root server itself). BigFix is simply serving out the file as a web service, so if a client cannot download the file, try downloading it from a browser.
I haven’t seen anyone mention it in this thread, but it is part of the Bare Metal setup recommendation (and I think there are some health check analyses that mention this)…but have you increased the Relay Cache sizes? For Bare Metal, you need to be sure to increase the Relay Cache Size - on all of the OSD Servers AND on every Relay in their upstream chain - to values large enough to accomodate WIM image downloads. From your same post, it looks like your WIM image size is 11,171,753,096 - which is far larger than the default 1 GB that a Relay can download & cache by default.
See https://bigfix.me/fixlet/details/515 for a fixlet to set the Relay Cache Sizes, and http://www-01.ibm.com/support/docview.wss?uid=swg21505905 and Legacy Communities - IBM TechXchange Community for more information.
…and now I’m seeing the same
On Relay 9.2.7.53 with server 9.5.1.
It appears that if I precache the image to the OSD server, it works for a couple of days until the image is aged-out of the Relay cache.
I’ll be trying the 9.5.2 upgrade next week.
…like @TheChad, I had upgraded the relay to 9.5.1 and then an in-place downgrade to 9.2.7. I haven’t gone through his process of clean removal of OSD and Relay.
Maybe drop back and punt?
- Uninstall OSD
- Run the Post Cleanup Task (134)
- Clear the Relay Cache
- Reinstall OSD using the latest (built-in download vs manually uploaded TPMfOSD)
- Re-Link bare metal profiles and full-sync it.
just a thought… start clean. using 9.5.2
I just did an in-place upgrade of the infrastructure, including the Relay service on the OSD server, to 9.5.2 this morning. On initial checkout it seems to be working as expected. When the image is requested by a client, I can watch it get cached to wwwrootbes/bfmirror/downloads/sha1 and subsequently download to the client.
No uninstall / reinstall / resync was needed.