Launching soon: The world's first vendor agnostic bug scrubLearn more & join waitlist
OPERATIONAL DEFECT DATABASE
...
...
XR VM cannot start on RP RP could be Active or Standby (nodes with single RP could be also affected)
no specific conditions
There is no workaround If the affected card is Active/Standby RP : Preferred method is whole router reload ('admin reload rack 0'). If the affected process is inst_agent on any non-RP card : Preferred method is affected card reload ('admin reload location all) Other recovery steps mentioned below can be tried if a whole router reload is not possible. But no guarantee that it would fix the problem. Also special attention need to be given to see both Active and standby Admin software packages are in sync (verify "admin show install active" - it should show current image for every node and make sure its in sync on all nodes prior to following the steps below). 1) If inst_mgr is the affected process we can attempt to restart inst_mgr in the Admin VM ('admin process restart inst_mgr location ') 2) If step 1 doesn’t work, we can attempt to shutdown inst_mgr in the Active RP and start the inst_mgr process in the Admin VM of the standby RP. Please make sure we do this in a controlled environment and also the admin software between active and standby cards are in sync (show install active) before we take this step. Else it can lead to bigger problems. 3) If step 2 also doesn’t help to fix the problem, whole router reload is needed to recover. 4) If inst_agent is the affected process we can attempt to restart inst_agent in the Admin VM ('admin process restart inst_agent location ) In case of image not being in sync (e.g. Between RP0 and RP1) one can try: - Remove Inactive RP from chassis - Reload chassis with single RP - Install removed RP back
sysadmin-vm:0_RP0# show sdr-operation Mon Mar 6 11:04:37.500 UTC+00:00 SDR: default-sdr Location Progress Status ------------------------------------- 0/RP1 62% In creation (being processed for install partition) <<<<<< VM creation process stuck on 62% sysadmin-vm:0_RP0# last observed logs for problem RP. 0/RP0/ADMIN0:Mar 8 20:03:16.845 : shelf_mgr[4238]: %INFRA-SHELF_MGR-6-CARD_SW_OPERATIONAL : Card: 0/RP1 software state going to Operational 0/RP0/ADMIN0:Mar 8 20:03:16.918 : esdma[5527]: %INFRA-ESDMA-6-ESD_CONN_FOUND : ESDMA found connection with esd at 0/RP1/RP-SW1 0/RP0/ADMIN0:Mar 8 20:03:16.918 : esdma[5527]: %INFRA-ESDMA-6-ESD_CONN_FOUND : ESDMA found connection with esd at 0/RP1/RP-SW1 0/RP1/ADMIN0:Mar 8 20:03:22.794 : inst_agent[4394]: %INFRA-INSTAGENT-4-XR_PART_PREP_REQ : Received SDR/XR partition request. Looking for available matching partition. If not found, new one will be created after copying relevant image and RPMs 0/RP1/ADMIN0:Mar 8 20:03:32.808 : inst_agent[4394]: %INFRA-INSTAGENT-2-SELF_TEST_FAILED : This node has failed verification of its installed software base. A new disk will be prepared. 0/RP0/ADMIN0:Mar 8 20:03:32.810 : inst_mgr[5548]: %INFRA-INSTMGR-5-SW_RESYNC : Node location 0/RP1 will resynchronize its software with the rest of the system. 0/RP1/ADMIN0:Mar 8 20:03:53.278 : i2c_server[4393]: %PLATFORM-I2C_SERVER-3-ACT2_AUTH_STATUS : ACT2 Authentication STATUS : ACT2 Authentication PASSED <<<<<<<<<<<<<<<<<<< nothing more for this RP (normally %INFRA-INSTAGENT-4-XR_PART_PREP_xxx and %INFRA-VM_MANAGER-4-INFO messages should appear saying that XR VM partition is prepared and XR VM is started)