Hello all, I'm trying to upgrade a pair of 9148 MDS switches from 6.2(21) to 6.2(23). I am getting an image verification error when running "sh install all impact": MDS-B# sh install all impact kickstart bootflash:m9100-s3ek9-kickstart-mz.6.2.23.bin system bootflash:m9100-s3ek9-mz.6.2.23.bin Installer will perform impact only check. Please wait. Verifying image bootflash:/m9100-s3ek9-kickstart-mz.6.2.23.bin for boot variable "kickstart". [# ] 0% -- FAIL. Pre-upgrade check failed. Return code 0x40930011 (Image verification failed). I've verified the MD5 on both images on the switch to ensure they copied over intact. MDS-B# sh file bootflash:m9100-s3ek9-mz.6.2.23.bin md5sum 0fec5ac4d2437d08398181812d054f73 MDS-B# sh file bootflash:m9100-s3ek9-kickstart-mz.6.2.23.bin md5sum a8cbaf73e3d6f2be93cf9c47f5900716 Both switches have plenty of space in bootflash, and were reloaded before attempting the install. Any thoughts? I don't see anything in the release notes for 6.2(23) or 6.2(21) that would explain why I might be seeing this issue.   ... View more

I am upgrading an HX240 cluster from 2.6(1a) to 2.6(1d), and the upgrade has completed successfully on 2 of the 4 nodes. One node fails after the bootstrap, VIB upgrade and reboot steps with the above error. I have re-attempted the upgrade multiple times to no avail. All nodes were initially installed at the same time, and there should not be any difference in the credentials between nodes. Is there any way to confirm what credentials it's trying to use, and whether the credentials match between this node and others? ... View more

I have 2 environments that were recently upgraded to 3.1(2e). After upgrading, I've found that logging into the CLI with local accounts is failing. It is also failing if I have the console auth set to local, and pass along my auth domain to log in via LDAP credentials, eg. using ucs-<authdomain>\<username>. The only way I can log into the CLI is if I change the native authentication to LDAP and log in with an AD account. My other environments that are running 3.1(2b) still work normally. Is this a known issue in 3.1(2e)? I'm not seeing it in the open caveats of the 3.1(2e) release notes nor am I finding anything in the bug tracker. It's not a significant issue for us so I haven't bothered to engage TAC yet...just wondering if others are seeing this issue as well. ... View more

Have hardware requirements for running Ent NFV on the C220 M4 been published? ... View more

We had a storage disruption event recently that appears to have flipped the root volume on our UCS Central VM to read-only. After rebooting the VM, I get an error message about needing to run a manual fsck due to filesystem errors on /dev/VolGroup00/LogVol00 (see attached screenshot), and I get dropped into a shell prompting for the root password. I can't seem to find any root credentials in the UCS documentation. Is there a procedure for repairing the filesystem and repairing the VM? ... View more

I have an environment where different teams own different aspects of the UCS environment. For example, operations is responsible for the environment and physical hardware, architecture is responsible for the configuration (policies, pools, profiles), and the network team is responsible for both LAN and SAN down to the blade. I'm trying to find a means of setting up call home profiles such that each team can get their relevant alerts, but I don't see how that's really possible within UCSM. Something that seems promising is alert groups, but I can't find any documentation on how/if they map to user roles or what alerts might fall into an alert group? The documentation meantions the alert groups, but without any description of what each alert group does or how they should be used. Some seem obvious (eg. Environmental) while others are vague (Syslog Port?). Any ideas? ... View more