After upgrading Splunk Enterprise to 9.4, Kvstore initiated in failure status and diagnosed that the issue was impacting searches due to the following error message when doing a simple search in a stand-alone instance.
Could not load lookup=LOOKUP-exc_host
Could not load lookup=LOOKUP-splunk_security_essentials
The following can be used to help diagnose Mongo Upgrade failure:
E STORAGE [initandlisten] WiredTiger error (-31802) [1739632215:112475][2157746:0x7f6315a43b40], connection: __log_open_verify, 925: unsupported WiredTiger file version: this build only supports versions up to 4, and the file is version 5: WT_ERROR: non-specific WiredTiger error Raw: [1739632215:112475][2157746:0x7f6315a43b40], connection: __log_open_verify, 925: unsupported WiredTiger file version: this build only supports versions up to 4, and the file is version 5: WT_ERROR: non-specific WiredTiger error
W STORAGE [initandlisten] Failed to start up WiredTiger under any compatibility version.
F STORAGE [initandlisten] Reason: -31802: WT_ERROR: non-specific WiredTiger error
F - [initandlisten] Fatal Assertion 28595 at src/mongo/db/storage/wiredtiger/wiredtiger_kv_engine.cpp 928
F - [initandlisten] \n\n***aborting after fassert() failure\n\n
INFO [mongod_upgrade::commands] UPGRADE_TO_4.4_SUCCESSFUL
INFO [mongod_upgrade::commands] Attempting to update status
INFO [mongod_upgrade::commands] Status updated successfully
INFO [mongod_upgrade] Waiting for other nodes in replica set to upgrade
DEBUG [mongod_upgrade::conditions] Upgraded count: 1
INFO [mongod_upgrade] All upgraded to 4.4, proceeding.
INFO [mongod_upgrade] Setting new FCV Version: 4.4
INFO [mongod_upgrade] FCV change successful: ()
INFO [mongod_upgrade] Received signal 15 - preparing to shut down
In the previous scenario, somehow Splunk decided to stop the upgrade after reaching 4.4 version. NOTE: a successful upgrade log should look like this:INFO [mongod_upgrade::commands] UPGRADE_TO_7.0_SUCCESSFUL
INFO [mongod_upgrade::commands] Attempting to update status
INFO [mongod_upgrade::commands] Status updated successfully
INFO [mongod_upgrade] Waiting for other nodes in replica set to upgrade
DEBUG [mongod_upgrade::conditions] Upgraded count: 1
INFO [mongod_upgrade] All upgraded to 7.0, proceeding.
INFO [mongod_upgrade] Setting new FCV Version: 7.0
INFO [mongod_upgrade] FCV change successful: ()
INFO [mongod_upgrade] Upgrades completed
INFO [mongod_upgrade] Waiting for completion
DEBUG [mongod_upgrade::conditions] Completed count: 1
DEBUG [mongod_upgrade::conditions] Hostname count: 1
INFO [mongod_upgrade] All completed
INFO [mongod_upgrade] Dropped migration metadata database.
INFO [mongod_upgrade::commands] Shutting down the database
WARN [mongod_upgrade] mongod failed to shut down before exiting: Kind: I/O error: unexpected end of file, labels: {}
To fix the error causes by the incomplete upgrade, perform the next steps:
[kvstore]
kvstoreUpgradeOnStartupEnabled = false
mv $SPLUNK_HOME/var/lib/splunk/kvstore/mongo $SPLUNK_HOME/var/lib/splunk/kvstore/mongo_bak
mv $SPLUNK_HOME/var/lib/splunk/kvstore/mongo_backup/mongo $SPLUNK_HOME/var/lib/splunk/kvstore/
$SPLUNK_HOME/bin/splunk start-standalone-upgrade kvstore -version 7.0 -dryRun true
$SPLUNK_HOME/bin/splunk start-standalone-upgrade kvstore -version 7.0
$SPLUNK_HOME/bin/splunk show standalone-kvupgrade-status
$SPLUNK_HOME/bin/splunk show kvstore-status --verbose
$SPLUNK_HOME/bin/splunk cmd mongod -version
Search Head Cluster Troubleshooting
In case steps above does not work by upgrading one by one in a Search Head Cluster Environment, follow these steps, in these ones is a must to take a backup of kvstore since next step is going to clean kvstore to force the upgrade in a fresh kvstore directory: Kvstore has stuck at starting stage for all the search heads in cluster.
Upgrade the KV store server version
Kvstore has stuck at starting stage for all the search heads in cluster.
| Revision | Publish Date | Comments |
|---|---|---|
1.0 |
29-Sep-2026
|
Initial Release |