]> git.uio.no Git - u/mrichter/AliRoot.git/blobdiff - SHUTTLE/AliShuttle.cxx
updating to new Shuttle processing logic
[u/mrichter/AliRoot.git] / SHUTTLE / AliShuttle.cxx
index a7f736ce6dfe3d075495358cb7ec41ef93c2ce24..f83a4f49f1f70209e2b4f4a107834c961295bbeb 100644 (file)
@@ -200,6 +200,7 @@ Bool_t AliShuttle::StoreLocally(const TString& localUri,
        //
        // returns 0 if fail, 1 otherwise
 
+
        if (fTestMode & kErrorStorage)
        {
                Log(fCurrentDetector, "StoreLocally - In TESTMODE - Simulating error while storing locally");
@@ -234,8 +235,11 @@ Bool_t AliShuttle::StoreLocally(const TString& localUri,
        if (!(AliCDBManager::Instance()->GetStorage(localUri))) {
                Log("SHUTTLE", Form("StoreLocally - Cannot activate local %s storage", cdbType));
        } else {
+               Int_t logLevel = AliLog::GetGlobalLogLevel();
+               AliLog::SetGlobalLogLevel(AliLog::kError);
                result = AliCDBManager::Instance()->GetStorage(localUri)
                                        ->Put(object, id, metaData);
+                AliLog::SetGlobalLogLevel((AliLog::EType_t)logLevel);
        }
 
        if(!result) {
@@ -243,6 +247,7 @@ Bool_t AliShuttle::StoreLocally(const TString& localUri,
                Log(fCurrentDetector, Form("StoreLocally - Can't store object <%s>!", id.ToString().Data()));
        }
 
+
        return result;
 }
 
@@ -865,7 +870,7 @@ const char* AliShuttle::GetRefFilePrefix(const char* base, const char* detector)
        //
 
        TString offDetStr(GetOfflineDetName(detector));
-       TString dir;
+       static TString dir;
        if (offDetStr == "ITS" || offDetStr == "MUON" || offDetStr == "PHOS")
        {
                dir.Form("%s/%s/%s", base, offDetStr.Data(), detector);
@@ -874,8 +879,6 @@ const char* AliShuttle::GetRefFilePrefix(const char* base, const char* detector)
        }
        
        return dir.Data();
-       
-
 }
 
 //______________________________________________________________________________________________
@@ -983,9 +986,14 @@ Bool_t AliShuttle::WriteShuttleStatus(AliShuttleStatus* status)
        fStatusEntry = new AliCDBEntry(status, id, new AliCDBMetaData);
        fStatusEntry->SetOwner(1);
 
+       Int_t logLevel = AliLog::GetGlobalLogLevel();
+        AliLog::SetGlobalLogLevel(AliLog::kError);
+
        UInt_t result = AliCDBManager::Instance()->GetStorage(fgkLocalCDB)->Put(fStatusEntry);
 
-       if (!result) {
+        AliLog::SetGlobalLogLevel((AliLog::EType_t)logLevel);
+
+       if (!result) {
                Log("SHUTTLE", Form("WriteShuttleStatus - Failed for %s, run %d",
                                                fCurrentDetector.Data(), run));
                return kFALSE;
@@ -1025,8 +1033,13 @@ void AliShuttle::UpdateShuttleStatus(AliShuttleStatus::Status newStatus, Bool_t
        status->SetStatus(newStatus);
        if (increaseCount) status->IncreaseCount();
 
+       Int_t logLevel = AliLog::GetGlobalLogLevel();
+        AliLog::SetGlobalLogLevel(AliLog::kError);
+
        AliCDBManager::Instance()->GetStorage(fgkLocalCDB)->Put(fStatusEntry);
 
+        AliLog::SetGlobalLogLevel((AliLog::EType_t)logLevel);
+
        SendMLDetInfo();
 }
 
@@ -1168,6 +1181,29 @@ Bool_t AliShuttle::ContinueProcessing()
                // Send mail to detector expert!
                Log("SHUTTLE", Form("ContinueProcessing - Sending mail to %s expert...", 
                                    fCurrentDetector.Data()));
+               // det experts in to
+               TString to="";
+               TIter *iterExperts = 0;
+               iterExperts = new TIter(fConfig->GetResponsibles(fCurrentDetector));
+               TObjString *anExpert=0;
+               while ((anExpert = (TObjString*) iterExperts->Next()))
+                       {
+                               to += Form("%s, \n", anExpert->GetName());
+                       }
+               delete iterExperts;
+               
+               if (to.Length() > 0)
+                       to.Remove(to.Length()-3);
+               AliDebug(2, Form("to: %s",to.Data()));
+
+               if (to.IsNull()) {
+                       Log("SHUTTLE", Form("List of %s responsibles not set!", fCurrentDetector.Data()));
+                       return kFALSE;
+               }
+
+               Log(fCurrentDetector.Data(), Form("ContinueProcessing - Sending mail to %s expert(s):", 
+                                   fCurrentDetector.Data()));
+               Log(fCurrentDetector.Data(), Form("\n%s", to.Data()));
                if (!SendMail(kPPEMail))
                        Log("SHUTTLE", Form("ContinueProcessing - Could not send mail to %s expert",
                                            fCurrentDetector.Data()));
@@ -1203,6 +1239,9 @@ void AliShuttle::SendMLRunInfo(const char* status)
                runType += fLogbookEntry->GetRunParameter("log");
                runType += ")";
        }
+       if (fLogbookEntry->GetDATestMode()){
+               runType += " (DATest)";
+       }
        TMonaLisaText  mlRunType("SHUTTLE_runtype", runType);
 
        TList mlList;
@@ -1319,21 +1358,27 @@ Bool_t AliShuttle::Process(AliShuttleLogbookEntry* entry)
        // Initialization
        Bool_t hasError = kFALSE;
 
-       // Set the CDB and Reference folders according to the year and LHC period
-       TString lhcPeriod(GetLHCPeriod());
-       if (lhcPeriod.Length() == 0) 
-       {
-               Log("SHUTTLE","Process - LHCPeriod not found in logbook!");
-               return 0; 
-       }       
-       
-       // build cdb paths (repeat each time, LHCperiod might have changed)
-       fgkMainCDB.Form("alien://folder=%s%d/%s/OCDB?user=alidaq?cacheFold=/tmp/OCDBCache", 
-                                       fConfig->GetAlienPath(), GetCurrentYear(), lhcPeriod.Data());
-       
-       fgkMainRefStorage.Form("alien://folder=%s%d/%s/Reference?user=alidaq?cacheFold=/tmp/OCDBCache", 
-                                       fConfig->GetAlienPath(), GetCurrentYear(), lhcPeriod.Data());
-       
+       // Set the CDB and Reference folders according to the year
+
+       // build cdb paths (repeat each time, run might be a DATest run)
+       if (!fLogbookEntry->GetDATestMode()){
+               fgkMainCDB.Form("alien://folder=%s%d/OCDB?user=alidaq?cacheFold=/tmp/OCDBCache", 
+                               fConfig->GetAlienPath(), GetCurrentYear());
+               
+               fgkMainRefStorage.Form("alien://folder=%s%d/Reference?user=alidaq?cacheFold=/tmp/OCDBCache", 
+                                      fConfig->GetAlienPath(), GetCurrentYear());
+       }
+       else {
+               fgkMainCDB.Form("alien://folder=%s%d/DATest/OCDB?user=alidaq?cacheFold=/tmp/OCDBCache",
+                               fConfig->GetAlienPath(), GetCurrentYear());
+               
+               fgkMainRefStorage.Form("alien://folder=%s%d/DATest/Reference?user=alidaq?cacheFold=/tmp/OCDBCache",
+                                      fConfig->GetAlienPath(), GetCurrentYear());
+       }
+
+       AliDebug(2,Form("Main CDB storage = %s",fgkMainCDB.Data()));
+       AliDebug(2,Form("Main Reference storage = %s",fgkMainRefStorage.Data()));
+
        // Loop on detectors in the configuration
        TIter iter(fConfig->GetDetectors());
        TObjString* aDetector = 0;
@@ -1892,7 +1937,7 @@ Bool_t AliShuttle::QueryShuttleLogbook(const char* whereClause,
        }
 
        // TODO Check field count!
-       const UInt_t nCols = 23;
+       const UInt_t nCols = 26;
        if (aResult->GetFieldCount() != (Int_t) nCols) {
                Log("SHUTTLE", "Invalid SQL result field number!");
                delete aResult;
@@ -1908,6 +1953,11 @@ Bool_t AliShuttle::QueryShuttleLogbook(const char* whereClause,
                if (!entry)
                        continue;
 
+               // DA test mode flag
+               TString daTestModeString(aRow->GetField(2), aRow->GetFieldLength(2)); // field 2 = DA test mode flag 
+               Bool_t daTestMode = (Bool_t)daTestModeString.Atoi();
+               entry->SetDATestMode(daTestMode);
+
                // loop on detectors
                for(UInt_t ii = 0; ii < nCols; ii++)
                        entry->SetDetectorStatus(aResult->GetFieldName(ii), aRow->GetField(ii));
@@ -1972,38 +2022,45 @@ AliShuttleLogbookEntry* AliShuttle::QueryRunParameters(Int_t run)
        UInt_t startTime = entry->GetStartTime();
        UInt_t endTime = entry->GetEndTime();
        Bool_t ecsSuccess = entry->GetECSSuccess();
-       
-       TString totEventsStr = entry->GetRunParameter("totalEvents");  
-       Int_t totEvents = totEventsStr.Atoi();
+       TString runType = entry->GetRunType();
+       TString tmpdaqstartTime = entry->GetRunParameter("DAQ_time_start");
+       UInt_t daqstartTime = tmpdaqstartTime.Atoi();
        
        UInt_t now = time(0);
-       // TODO make this a configuration parameter
        Int_t dcsDelay = fConfig->GetDCSDelay()+fConfig->GetDCSQueryOffset();
-       
-       // runs are accepted if they have ecsSuccess set or more than 1 event
-       if (startTime != 0 && endTime != 0 && endTime > startTime && (totEvents > 1 || ecsSuccess) && (endTime < now - dcsDelay))
-       {
-               if (ecsSuccess == kFALSE)
-                       Log("SHUTTLE", Form("Processing run %d although in status ECS failure, Reason: %s", run, entry->GetRunParameter("eor_reason")));
-               return entry;
-       }
 
        Bool_t skip = kFALSE;
                                
-       if (endTime != 0 && endTime >= now - dcsDelay)
-       {
-               Log("SHUTTLE", Form("Skipping run %d for now, because DCS buffer time is not yet expired", run));
-       }
-       else if (totEvents <= 1) 
-       {
-               Log("SHUTTLE", Form("QueryRunParameters - Run %d has 1 event or less - Skipping!", run));
+       // runs are processed if
+       //   a) runType is PHYSICS and ecsSuccess is set
+       //   b) runType is not PHYSICS and (ecsSuccess is set or DAQ_time_start is non-0)
+       // effectively this means that all runs are processed that started properly (ecsSucess behaviour is different for PHYSICS and non-PHYSICS runs (check with ECS!)
+       if (startTime != 0 && endTime != 0 && endTime > startTime) { 
+         if (endTime >= now - dcsDelay) {
+           Log("SHUTTLE", Form("Skipping run %d for now, because DCS buffer time is not yet expired", run));
+         } else {
+           if (runType == "PHYSICS") {
+             if (ecsSuccess) {
+               return entry;
+             } else {
+               Log("SHUTTLE", Form("QueryRunParameters - Run type for run %d is PHYSICS but ECS success flag not set (Reason = %s) - Skipping!", run, entry->GetRunParameter("eor_reason")));
                skip = kTRUE;
-       }
-       else
-       {
-               Log("SHUTTLE", Form("QueryRunParameters - Invalid parameters for Run %d: "
-                       "startTime = %d, endTime = %d. Skipping (Shuttle won't be marked as DONE)!",
-                       run, startTime, endTime));
+             } 
+           } else {
+             if (ecsSuccess || daqstartTime > 0) {
+               if (ecsSuccess == kFALSE)
+                 Log("SHUTTLE", Form("Processing run %d although in status ECS failure (Reason: %s), since run type != PHYSICS and DAQ_time_start != 0", run, entry->GetRunParameter("eor_reason")));
+               return entry;
+             } else {
+               Log("SHUTTLE", Form("QueryRunParameters - Run type for run %d is %s, ECS success flag was not set (Reason = %s) and DAQ_time_start was NULL - Skipping!", run, runType.Data(), entry->GetRunParameter("eor_reason")));
+               skip = kTRUE;
+             }
+           }
+         }
+       } else {
+         Log("SHUTTLE", Form("QueryRunParameters - Invalid parameters for Run %d: "
+                             "startTime = %d, endTime = %d. Skipping (Shuttle won't be marked as DONE)!",
+                             run, startTime, endTime));
        }
        
        if (skip)
@@ -2340,7 +2397,7 @@ TList* AliShuttle::GetFileSources(Int_t system, const char* detector, const char
                return NULL;
        }
 
-       TString sourceName = 0;
+       TString sourceName = "";
        if (system == kDAQ)
        {
                sourceName = "DAQsource";
@@ -2419,7 +2476,7 @@ TList* AliShuttle::GetFileIDs(Int_t system, const char* detector, const char* so
                return NULL;
        }
 
-       TString sourceName = 0;
+       TString sourceName = "";
        if (system == kDAQ)
        {
                sourceName = "DAQsource";
@@ -2483,7 +2540,20 @@ Bool_t AliShuttle::Connect(Int_t system)
        //
 
        // check connection: if already connected return
-       if(fServer[system] && fServer[system]->IsConnected()) return kTRUE;
+
+       if(fServer[system] && fServer[system]->IsConnected()) {
+               // ping the server              
+               if (fServer[system]->PingVerify()==kTRUE){ // connection is still alive
+                       return kTRUE;
+               }               
+               else{
+                       AliWarning(Form("Connection got lost to FXS database for %s. Closing and reconnecting.",
+                                       AliShuttleInterface::GetSystemName(system)));
+                       fServer[system]->Close();
+                       delete fServer[system];
+                       fServer[system] = 0x0;
+               }
+       }
 
        TString dbHost, dbUser, dbPass, dbName;
 
@@ -2627,6 +2697,8 @@ Bool_t AliShuttle::UpdateTableSkippedCase(const char* detector)
 
        Bool_t result = kTRUE;
 
+       TString detName(detector);
+
        for (UInt_t system=0; system<3; system++)
        {
 
@@ -2644,10 +2716,10 @@ Bool_t AliShuttle::UpdateTableSkippedCase(const char* detector)
                TIter iter(&fFXSlist[system]);
                        
                TString whereClause;
-               if (detector == "ALL") whereClause = Form("where run=%d and time_processed IS NULL;",GetCurrentRun());
+               if (detName == "ALL") whereClause = Form("where run=%d and time_processed IS NULL;",GetCurrentRun());
                else whereClause = Form("where run=%d and detector=\"%s\" and time_processed IS NULL;",GetCurrentRun(), detector);
 
-               Log("SHUTTLE",Form(" whereClause = %s ",whereClause.Data()));
+               //Log("SHUTTLE",Form(" whereClause = %s ",whereClause.Data()));
 
                TString sqlQuery = Form("update %s set time_processed=%d %s", fConfig->GetFXSdbTable(system),
                                        now.GetSec(), whereClause.Data());
@@ -2813,16 +2885,6 @@ UInt_t AliShuttle::GetCurrentEndTime() const
 
        return fLogbookEntry ? fLogbookEntry->GetEndTime() : 0;
 }
-//______________________________________________________________________________________________
-UInt_t AliShuttle::GetCurrentTimeCreated() const
-{
-       //
-       // get current end time from logbook entry
-       //
-
-       return fLogbookEntry ? fLogbookEntry->GetTimeCreated() : 0;
-}
-
 //______________________________________________________________________________________________
 UInt_t AliShuttle::GetCurrentYear() const
 {
@@ -3154,7 +3216,7 @@ Bool_t AliShuttle::SendMail(EMailTarget target, Int_t system)
 
        if (target == kDCSEMail || target == kFXSEMail) {
                if (!fFirstProcessing)
-               return kTRUE;
+                       return kTRUE;
        }
 
        Int_t runMode = (Int_t)fConfig->GetRunMode();
@@ -3261,6 +3323,7 @@ Bool_t AliShuttle::SendMail(EMailTarget target, Int_t system)
                body = Form("Dear  %s FXS experts, \n\n",sys.Data());
                body += Form("SHUTTLE couldn\'t retrieve data from the FXS for detector %s "
                             "in run %d!!\n\n", fCurrentDetector.Data(), GetCurrentRun());
+               body += Form("The contacted server was:\nDB: %s\nFXS:%s\n\n", fConfig->GetFXSdbHost(system), fConfig->GetFXSHost(system));
        }
        else {
                subject = Form("%s %s Shuttle preprocessor FAILED in run %d (run type = %s)!",
@@ -3289,10 +3352,10 @@ Bool_t AliShuttle::SendMail(EMailTarget target, Int_t system)
        
        
        body += Form("Find the %s log for the current run on \n\n"
-               "\thttp://pcalishuttle01.cern.ch:8880/%s/%d/%d/%s.log \n\n", 
+               "\thttp://pcalishuttle02.cern.ch/%s/%d/%d/%s.log \n\n", 
                     fCurrentDetector.Data(), logFolder.Data(), GetCurrentRun()/10000,  
                                GetCurrentRun(), fCurrentDetector.Data());
-       body += Form("The last 10 lines of %s log file are following:\n\n", fCurrentDetector.Data());
+       body += Form("The last 15 lines of %s log file are following:\n\n", fCurrentDetector.Data());
 
        AliDebug(2, Form("Body begin: %s", body.Data()));
 
@@ -3302,7 +3365,7 @@ Bool_t AliShuttle::SendMail(EMailTarget target, Int_t system)
 
        TString logFileName = Form("%s/%d/%d/%s.log", GetShuttleLogDir(), 
                GetCurrentRun()/10000, GetCurrentRun(), fCurrentDetector.Data());
-       TString tailCommand = Form("tail -n 10 %s >> %s", logFileName.Data(), bodyFileName.Data());
+       TString tailCommand = Form("tail -n 15 %s >> %s", logFileName.Data(), bodyFileName.Data());
        if (gSystem->Exec(tailCommand.Data()))
        {
                mailBody << Form("%s log file not found ...\n\n", fCurrentDetector.Data());
@@ -3331,7 +3394,6 @@ Bool_t AliShuttle::SendMail(EMailTarget target, Int_t system)
 
        return result == 0;
 }
-
 //______________________________________________________________________________________________
 const char* AliShuttle::GetRunType()
 {
@@ -3394,7 +3456,7 @@ const char* AliShuttle::GetTriggerConfiguration()
        
        if (result->GetRowCount() == 0)
        {
-               Log("SHUTTLE", "ERROR: Trigger configuration not found in logbook_trigger_config");
+               Log("SHUTTLE", "WARNING: Trigger configuration not found in logbook_trigger_config");
                delete result;
                return 0;
        }
@@ -3421,6 +3483,100 @@ const char* AliShuttle::GetTriggerConfiguration()
        return triggerConfig;
 }
 
+//______________________________________________________________________________________________
+const char* AliShuttle::GetCTPTimeParams()
+{
+       // Receives the CTP time parameters from the DAQ logbook for the current run
+       
+       // check connection, if needed reconnect
+       if (!Connect(3)) 
+               return 0;
+
+       TString sqlQuery;
+       sqlQuery.Form("SELECT alignmentFile FROM logbook_trigger_config WHERE run = %d", GetCurrentRun());
+       TSQLResult* result = fServer[3]->Query(sqlQuery);
+       if (!result)
+       {
+               Log("SHUTTLE", Form("ERROR: Can't execute query <%s>!", sqlQuery.Data()));
+               return 0;
+       }
+       
+       if (result->GetRowCount() == 0)
+       {
+               Log("SHUTTLE", "WARNING: CTP time params not found in logbook_trigger_config");
+               delete result;
+               return 0;
+       }
+       
+       TSQLRow* row = result->Next();
+       if (!row)
+       {
+               Log("SHUTTLE", "ERROR: Could not receive logbook_trigger_config data");
+               delete result;
+               return 0;
+       }
+
+       // static, so that pointer remains valid when it is returned to the calling class       
+       static TString triggerTimeParams(row->GetField(0));
+       
+       delete row;
+       row = 0;
+       
+       delete result;
+       result = 0;
+       
+       Log("SHUTTLE", Form("Found trigger time parameters: %s", triggerTimeParams.Data()));
+       
+       return triggerTimeParams;
+}
+
+//______________________________________________________________________________________________
+const char* AliShuttle::GetTriggerDetectorMask()
+{
+       // Receives the trigger detector mask from DAQ logbook
+       
+       // check connection, if needed reconnect
+       if (!Connect(3)) 
+               return 0;
+
+       TString sqlQuery;
+       sqlQuery.Form("SELECT BIN(BIT_OR(inputDetectorMask)) from logbook_trigger_clusters WHERE run = %d;", GetCurrentRun());
+       TSQLResult* result = fServer[3]->Query(sqlQuery);
+       if (!result)
+       {
+               Log("SHUTTLE", Form("ERROR: Can't execute query <%s>!", sqlQuery.Data()));
+               return 0;
+       }
+       
+       if (result->GetRowCount() == 0)
+       {
+               Log("SHUTTLE", "ERROR: Trigger Detector Mask not found in logbook_trigger_clusters");
+               delete result;
+               return 0;
+       }
+       
+       TSQLRow* row = result->Next();
+       if (!row)
+       {
+               Log("SHUTTLE", "ERROR: Could not receive logbook_trigger_clusters data");
+               delete result;
+               return 0;
+       }
+
+       // static, so that pointer remains valid when it is returned to the calling class       
+       static TString triggerDetectorMask(row->GetField(0));
+       
+       delete row;
+       row = 0;
+       
+       delete result;
+       result = 0;
+       
+       Log("SHUTTLE", Form("Found Trigger Detector Mask: %s", triggerDetectorMask.Data()));
+       
+       return triggerDetectorMask;
+}
+
 //______________________________________________________________________________________________
 void AliShuttle::SetShuttleTempDir(const char* tmpDir)
 {
@@ -3455,7 +3611,7 @@ Bool_t AliShuttle::TouchFile()
        }
 
        TString dir;
-       dir.Form("%s%d/%s/SHUTTLE_DONE", fConfig->GetAlienPath(), GetCurrentYear(), GetLHCPeriod());
+       dir.Form("%s%d/SHUTTLE_DONE", fConfig->GetAlienPath(), GetCurrentYear());
        // checking whether directory for touch command exists
        TString commandLs;
        commandLs.Form("ls %s",dir.Data());
@@ -3472,7 +3628,7 @@ Bool_t AliShuttle::TouchFile()
                return kFALSE;
        }
        TObjString *valueLsPath = dynamic_cast<TObjString*>(mapLs->GetValue("path"));
-       if (!valueLsPath || (TString)(valueLsPath->GetString()).CompareTo(dir)!=1){ 
+       if (!valueLsPath || (valueLsPath->GetString()).CompareTo(dir)!=1){ 
                Log("SHUTTLE",Form("No directory %s found, creating it",dir.Data()));
 
                // creating the directory
@@ -3527,16 +3683,16 @@ Bool_t AliShuttle::TouchFile()
        return kTRUE;
 }
 //______________________________________________________________________________________________
-const UInt_t AliShuttle::GetStartTimeDCSQuery()
+UInt_t AliShuttle::GetStartTimeDCSQuery()
 {
        // Return Start Time for the DCS query
        //
        // The call is delegated to AliShuttleInterface
 
-       return GetCurrentTimeCreated()-fConfig->GetDCSQueryOffset();
+       return GetCurrentStartTime()-fConfig->GetDCSQueryOffset();
 }
 //______________________________________________________________________________________________
-const UInt_t AliShuttle::GetEndTimeDCSQuery()
+UInt_t AliShuttle::GetEndTimeDCSQuery()
 {
        // Return End Time for the DCS query
        //
@@ -3545,4 +3701,3 @@ const UInt_t AliShuttle::GetEndTimeDCSQuery()
        return GetCurrentEndTime()+fConfig->GetDCSQueryOffset();
 }
 
-