Errors In QV Event Log : PhoenixThread: Stats request stuck report end PhoenixThread: Not waiting or awaiting

What do PhoenixThread: Stats request stuck report end;  PhoenixThread: Not waiting or awaiting mean and other PhoenixThread errors mean when seen in the QlikView Server Events log?

One reason may be a bug in 12.0 and 12.10 where sometimes, overworked PGO file access may result in QlikView Server restarts due to watchdog thread "hanging".

If the .PGO files in the ROOT folder become inaccessible for a period of time, particularly in a clustered environment, there is a possibility of seeing Phoenix Thread restarts in the QlikView Server Event logs.

If .PGO files become overloaded (.PGO file access and read failures in QlikView Event log), it is possible the QlikView Server Watchdog thread (used to help find problems such as dead-locks and is regularly inserted to address a lock) will hang, trying to manage the lock.  QlikView Server notes the current timestamp of the server Watchdog.thread  Another thread - the Phoenix thread,  checks the timestamp and if too much time has passed, QlikView Server will restart.  

May see the following types of PhoenixThread errors:

PhoenixThread: Stats request stuck report begin

PhoenixThread: Waiting on monitor CMon<slvl=1000 rlvl=1 age=-1 cu=<Fiber(0000000D2AEC0820)<id=391 #susp=0 pnd=0 lpnd=6>CAnyRequest(0x000000177CBA4C48)<hasQv=1 hasFiber=1 bprio=1>CQvReq(0x000000177CBA4C20)<f=Global::CustomProtocolRequest pUnk=0x000000004E618C20 hasFiber=1 onbr=113907675 strd=1 fin=0 isVal=0 ex=-64 eex=0>> #aw=0 #w=34>

PhoenixThread: ExecContext (depth 2): Fiber(00000010915A7020)<id=340 #susp=1 pnd=0 lpnd=6>CAnyRequest(0x0000000A8053E048)<hasQv=1 hasFiber=1 bprio=1>CQvReq(0x0000000A8053E020)<f=Global::OpenDocEx pUnk=0x0000000C6F4C0820 hasFiber=1 onbr=113836745 strd=1 fin=0 isVal=0 ex=0 eex=0>

PhoenixThread: Stats request stuck report end

PhoenixThread: Not waiting or awaiting

 

 

There is a bug in 12.0 and 12.10 where at times, overworked PGO access can result in QV server restarts due to watchdog thread becomming "hung". 

Jira issue QV-8813  "QV-PGOs is producing unstable results or failing when run on VM" 

Ensure ROOT folder access is continuous and without delay, Anti-Virus scan exceptions are set for all  QlikView folders including .pgo file location.

Perform due diligence upon the QlikView Server ROOT folder.  Ensure timestamps are current for all .PGO files, and confirm files in the Backup folders on the QLikView servers are in sync.

Clean up the ROOT folder of any old .pgo files with no extensions (e.g. CalD, or IniD without .pgo extensions or symbols embedded in the name.  Best practice is for the ROOT folder to only house the 5 .PGO files.

Fix for issue will be in 12.20 track.