One of the two cluster FIs have a bootflash capacity-exceed warning

[Hele Du]

Hi everybody,

Could someone can solve this or show me what files in the bootflash?

I found some documents on Cisco site ,  the bootflash utilization should equal.

One of them is 30% and another is almost 70%

[Walter Dey]

Hi Hele

Most likely you store many UCS versions ! just go and do a cleanup, delete old versions and you will recover quickly disk space.

Walter.

[Hele Du]

Hi Walter,

I used the show tech-support command to figure out what files in the bootflash.

There are dir bootflash for FIA and FIB resepctively.

I couldn`t found any file esle.

Is there have any hidden files in bootfflash?

FIa:

`dir bootflash:`
         14    Feb 06 18:18:32 2015  TZ_DATA.DAT
         10    May 13 19:10:13 2011  bootflash
   21584778    Feb 06 17:41:25 2015  chassis.img
   34946282    Feb 06 17:41:26 2015  chassis2.img
         20    Feb 06 17:41:45 2015  cores
       4096    Apr 18 00:39:36 2014  distributables/
       4096    Apr 18 00:39:36 2014  distributables_hdr/
   31976960    Feb 06 17:41:26 2015  fexth.bin
         52    May 13 11:41:59 2011  initial_setup.log
       4096    Feb 04 21:48:54 2015  installables/
       4096    May 13 19:12:07 2011  license/
      49152    Oct 21 02:50:21 2012  lost+found/
       6886    Feb 06 17:42:08 2015  mts.log
         56    Oct 21 02:30:58 2012  nuova-sim-mgmt-nsg.0.1.0.001.bin
         29    Feb 06 17:41:45 2015  pnuos
       4096    Oct 20 22:03:57 2012  received/
       4096    Dec 27 08:46:38 2012  sysdebug/
       4096    May 13 19:42:24 2011  techsupport/
   21581312    Feb 16 07:59:22 2011  ucs-6100-k9-kickstart.4.1.3.N2.1.3i.bin
  157634819    Feb 16 08:00:02 2011  ucs-6100-k9-system.4.1.3.N2.1.3i.bin
       4096    Feb 16 08:04:59 2011  vdc_2/
       4096    Feb 16 08:04:59 2011  vdc_3/
       4096    Feb 16 08:04:59 2011  vdc_4/
       4096    May 13 19:45:16 2011  vms_cert_store/

Usage for bootflash://sup-local
 2692239360 bytes used
 6386520064 bytes free
 9078759424 bytes total
 

FIb:

`dir bootflash:`
         14    Feb 06 17:26:38 2015  TZ_DATA.DAT
         10    May 13 19:12:00 2011  bootflash
   21584778    Feb 06 17:22:21 2015  chassis.img
   34946282    Feb 06 17:22:22 2015  chassis2.img
         20    Feb 06 17:22:43 2015  cores
       4096    Apr 18 00:39:36 2014  distributables/
       4096    Apr 18 00:39:36 2014  distributables_hdr/
   31976960    Feb 06 17:22:22 2015  fexth.bin
         52    May 13 14:52:52 2011  initial_setup.log
       4096    Feb 04 21:48:54 2015  installables/
       4096    May 13 19:45:08 2011  license/
      49152    Oct 21 02:16:56 2012  lost+found/
     102419    Feb 06 17:23:03 2015  mts.log
         56    Oct 21 02:10:31 2012  nuova-sim-mgmt-nsg.0.1.0.001.bin
         29    Feb 06 17:22:43 2015  pnuos
       4096    May 13 19:15:03 2011  received/
       4096    Dec 27 08:23:19 2012  sysdebug/
       4096    May 13 19:30:22 2011  techsupport/
   21581312    Feb 16 08:29:51 2011  ucs-6100-k9-kickstart.4.1.3.N2.1.3i.bin
  157634819    Feb 16 08:30:32 2011  ucs-6100-k9-system.4.1.3.N2.1.3i.bin
       4096    Feb 16 08:35:58 2011  vdc_2/
       4096    Feb 16 08:35:58 2011  vdc_3/
       4096    Feb 16 08:35:58 2011  vdc_4/
       4096    May 13 14:54:37 2011  vms_cert_store/

Usage for bootflash://sup-local
 6646411264 bytes used
 2432348160 bytes free
 9078759424 bytes total

[Walter Dey]

Could it be, that there are tar files from show tech which were not deleted ?

they are under workspace:///techsupport/.....

Did you have a TAC case and installed any plugins ?

https://supportforums.cisco.com/document/66296/how-read-ucs-b-series-tech-support-files-ucsm-detail

[Hele Du]

HI Walter,

Thank you for your reply.

As far as I know, there are three partition in FI--bootflash; workspace and opt respectively.

In this scenario only bootflash exceed 70%, it wasn`t generated a warning message that workspace has been exceed 70% event there are lots show techsupport in workspace partition.

Here is a show file in FIb, I thought the workspace has enough space.

cqu-ucs-B(local-mgmt)# ls workspace:/

 1   16 Feb 16 08:54:04 2011 cores
 2 1024 May 13 19:12:01 2011 debug_plugin/
 1   31 Feb 16 08:54:05 2011 diagnostics
 2 1024 Feb 16 08:52:01 2011 lost+found/
 2 1024 Feb 16 08:52:54 2011 techsupport/

Usage for workspace://
291376128 bytes total
8427520 bytes used
267902976 bytes free
cqu-ucs-B(local-mgmt)#

Do you have any other idea for this?

[Walter Dey]

Did you check, if there are any core dumps around ?

[Hele Du]

I am sorry , I have no idea how to check the core dump.

I had checked many  documents, the only thing is the below:

If Cisco UCS Manager reports that the bootflash is out of space, delete obsolete bundles on the Packages tab to
free up space. To view the available space in bootflash, navigate to the fabric interconnect on the Equipment tab
and expand the Local Storage Information area on the General tab.

[Walter Dey · Accepted solution]

Hi Hele

See attachment, how you can find core dump files, and show tech .tar files. From there, you can also delete them.

Walter.

[Hele Du]

HI Walter,

It is really my appreciate!

Its a lots of core dump files in there, it will be fine if I delete them?

[Walter Dey]

Check the dates; maybe I would copy the 2 recent ones off the UCS fabric interconnect; then delete them all.

This however makes only sense, if you would like to open a Cisco TAC case; they would then analyze what the problem really was.

Good luck

Walter.

[Hele Du]

Yes , I think the TAC will be better for this ,but but but this UCS is out of services.

Haha, whatever, this issue has been fixed.

Thank you Walter.

[Keny Perez]

Check the Equipment>Equipment>Firmware Management? Packages and share the screenshot here.

-Kenny

[Walter Dey]

Has been done already in a previous post; see above !

[Keny Perez]

Oh I hadnt seen it... so he is running 2.0.4!  I think there were a couple of bugs for this, I just cant find them... better open a TAC case to dig deeper.

-Kenny