Title: | Archive/Backup |
Moderator: | COOKIE::MHUA IG |
Created: | Wed Sep 08 1993 |
Last Modified: | Fri Jun 06 1997 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 479 |
Total number of notes: | 2283 |
Hi all, Customer is having problems when using the 2.1 GUI. When he clicks to enter a save request it fails on him. This consistently reproducable. We have had motif people looking at this to see if they can spot anything strange going on. Here is what we have gathered. We had him do a set host 0 and turn on watch file. This is what we found: %XQP, Thread #0, Read attributes: Record attributes ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Owner UIC ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Header 1 accessibility ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: File protection ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: File access level ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: ACL length ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Access mode ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Journal flags ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: RU active ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Statistics block ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Find ACE by type ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: Record attributes ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Read attributes: User file characteristics ABS$POLICY_CONFIG.DAT;1 (6188,10,0)<CR> %XQP, Thread #0, Access ABS$POLICY_CONFIG.DAT;1 (6188,10,0) Status: 00000001<CR> Exception Module Line Exception Reraised by: PABUI_XWINDOW 848 Exception Reraised by: PABUI_MAINWINDOW 5028 exception PABUI_MAINWINDOW 5137 %XQP, Thread #0, FIB contents:<CR> 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000<CR> 00000000 00000000 00000000 00030000 00000000 00000000 00000000 00000000<CR> 00000000 00000000 00000000 00000000<CR> %XQP, Thread #0, FIB contents:<CR> 00000000 00031B1E 00000000 00000000 00000000 00000000 00000000 00000000<CR> Thinking this may have something to do with the ABS$POLICY_CONFIG.DAT file we got this from him: ! ! This is the configuration file for the abs_policy_engine ! for version 1.0 this is not dynamic ! ABS$AUDIT_ALL = FALSE ABS$COLLECT_STATISTICS = TRUE ABS$DEBUG_LOG = FALSE ! ABS$AUDIT_CREATE_ARCHIVE_CLASS = FALSE ABS$AUDIT_SET_ARCHIVE_CLASS = FALSE ABS$AUDIT_SHOW_ARCHIVE_CLASS = FALSE ABS$AUDIT_DELETE_ARCHIVE_CLASS = FALSE ! ABS$AUDIT_CREATE_ARCHIVE_REQUEST = TRUE ABS$AUDIT_SET_ARCHIVE_REQUEST = FALSE ABS$AUDIT_SHOW_ARCHIVE_REQUEST = FALSE ABS$AUDIT_DELETE_ARCHIVE_REQUEST = TRUE ! ABS$AUDIT_CREATE_RESTORE_REQUEST = TRUE ABS$AUDIT_SET_RESTORE_REQUEST = FALSE ABS$AUDIT_SHOW_RESTORE_REQUEST = FALSE ABS$AUDIT_DELETE_RESTORE_REQUEST = TRUE ! ABS$AUDIT_CREATE_EXEC_ENV = FALSE ABS$AUDIT_SET_EXEC_ENV = FALSE ABS$AUDIT_SHOW_EXEC_ENV = FALSE ABS$AUDIT_DELETE_EXEC_ENV = FALSE ! ABS$AUDIT_CLIENT = FALSE ! ABS$MAX_SHUTDOWN_WAIT = 5 ! ! client server image match checking ! ABS$IMAGE_VERSION_CHECK_LEVEL = MODERATE ! ! Net work object name / base name for audit log file ! ABS$NET_OBJECT_NAME = ABS$POLICY ! ! ! List of nodes for Policy Engine ! (appended in KITINSTAL at installation) ! ABS$POLICY_ENGINE_LOCATION = A4100:: There are some differences between our config file and theirs but attempts to use their config file on our machine have succeeded and we could not reproduce the problem. I'm having him try the LUI to see if it can create a SAVE REQUEST or if it crashes with the same problem. This is a VMS 6.2 machine...I believe AXP. Any ideas, Thanks... Ted
T.R | Title | User | Personal Name | Date | Lines |
---|---|---|---|---|---|
398.1 | Need more info | COOKIE::HELLWIG | Wed Mar 12 1997 10:55 | 17 | |
Couple of questions: - Version of Motif? - Version of VMS? - There are a couple ways of creating a save request. Does the gui fail when he pushes the save button at the botton, or when he uses the Create->Save Request pulldown option? It will also be good information to know if LUI works for them. The ABS$POLICY_CONFIG.DAT probably doesn't have anything to do with this. This is read in during the GUI startup, but the GUI itself doesn't do anything with the values. The information in the policy configuration file is for the policy engine startup, message leveling, etc. con Kim | |||||
398.3 | CSC32::V_HEINICKE | Wed Mar 12 1997 13:25 | 35 | ||
Couple of questions: - Version of Motif? V1.2-4a - Version of VMS? V6.2-1h3 - There are a couple ways of creating a save request. Does the gui fail when he pushes the save button at the botton, or when he uses the Create->Save Request pulldown option? It fails both ways. The Lui did not crash. He said he exercised it pretty hard. The only piece that did not work is when he did a save, he received the following messages: coordinator: sysput service failed on a TLE coordinator: facility ABS:abs_sysput_failed_tle, sys$put service failed on a TLE coordinator: line=59, file=RESD$:[SRC]coordinator.c;1 This sequence repeated about 5 times naming a different file each time. He is applying a patch for MOTIF that addresses an issue whereby when DW-Motif goes inot a fail safe condition, the login box does not reappear. Actually, he has applied the patch but cannot reboot the system so he has to wait until a reboot can be scheduled - he has a call into the Motif team to see if the is really needed. Thanks, Victoria | |||||
398.4 | Log needed | COOKIE::MHUA | Thu Mar 13 1997 08:55 | 10 | |
Hi, Can you post the whole save log from the failed run? The information is partial and the line number 59 is a comment line (it cannot singal error, you see). We need the whole diag dump to see how things failed. Masami | |||||
398.5 | Upgrade from V2.0? | COOKIE::HELLWIG | Thu Mar 13 1997 09:57 | 18 | |
Something else to look at: Did the customer upgrade from V2.0? If so, make sure the catalogs have been converted to the new format. If this was an upgrade from V2.0 they need to modify the following file. ABS$SYSTEM:CONVERT_CATALOG_V21.COM OLD: $ Convert/FDL=ABS$SYSTEM:TLE_BRIEF.FDL - ABS_CATALOG:'P1'_BTLE.DAT ABS_CATALOG:'P1'_BTLE.DAT NEW: $ Convert/FDL=ABS$SYSTEM:TLE_BRIEF.FDL/PAD=" " - ABS_CATALOG:'P1'_BTLE.DAT ABS_CATALOG:'P1'_BTLE.DAT Kim | |||||
398.6 | Tried the catalog conversion as well as creating new catalog - still having problem | CSC32::V_HEINICKE | Thu Mar 13 1997 14:40 | 27 | |
Hi, We did the convert on the catalog as per Kim's request. We also did the following: create/fdl=abs$system:aoe_brief.fdl abs_catalog:abs_cat_staged_baoe.dat create/fdl=abs$system:aoe_instance_brief.fdl abs_catalog:abs_cat_staged_baoe_ins nc.dat create/fdl=abs$system:tle_brief.fdl abs_catalog:abs_cat_staged_btle.dat The GUI save request dumps him out completely - the Lui returns the error I reported earlier. I had him do a 'set host 0/log=craig.2'. He is sending me the log file. I will put it in csc32::Dumps:[v_heinicke]. There will be a Craig.log and a Craig2.log with W:RE on it. I will try and remove the PABUI_SESSION.log file as it was created on FEB 26th and does not relate to this problem. No new files were created. Hope this helps. Thanks, Victoria | |||||
398.7 | Comments on craig1.txt | COOKIE::HELLWIG | Mon Mar 17 1997 12:42 | 112 | |
I did not see either craig.log or craig2.log, but I did find a craig.txt and a craig1.txt. I've got some comments on craig1.txt Based upon the information in craig1.txt, it looks like the installation did not work properly. Were there any failures during the installation? >Initializing Save Request SYSTEM_26-FEB-1997_15_50_03_38 >Storage Class SYSTEM_BACKUPS does not exist. Please choose a known Storage Clas >s. >Error is: Storage class not found >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > Line = 177, File = RESD$:[SRC]API_C_SHOW_ARCH_CLASS.C;1 >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > Line = 112, File = RESD$:[SRC]API_S_SHOW_ARCH_CLASS.C;1 >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > Line = 318, File = RESD$:[SRC]ARCHIVE_CLASS_READ.C;1 >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > Line = 562, File = RESD$:[SRC]ARCHIVE_CLASS_READ.C;1 >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > Line = 1583, File = RESD$:[SRC]ARCHIVE_CLASS_READ.C;1 >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > %RMS-E-RNF, record not found >Facility ABS: ABS_ARCH_CLASS_NOT_FOUND, Storage class not found > SQL error code = 100, Line = 2015, File = RESD$:[SRC]ARCHIVE_CLASS_READ.C;1 The storage class used by the save request is not found. Have them do a "$ABS SHOW STORAGE foo" using the storage class name referenced in the save request. If the storage class is not there, have them run the ABS$SYSTEM:EPCOT_DB_INIT.EXE program. > Submission of Save Request SYSTEM_26-FEB-1997_15_50_03_38 failed for the follow >ing reason: >Execution env not found >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 705, File = RESD$:[SRC]API_C_CRE_ARCH_REQ.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 550, File = RESD$:[SRC]API_C_CRE_ARCH_REQ.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 297, File = RESD$:[SRC]API_S_CRE_ARCH_REQ.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 112, File = RESD$:[SRC]API_S_SHOW_EXEC_ENVIR.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 344, File = RESD$:[SRC]EXECUTION_ENV_READ.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 583, File = RESD$:[SRC]EXECUTION_ENV_READ.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > Line = 1713, File = RESD$:[SRC]EXECUTION_ENV_READ.C;1 >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > %RMS-E-RNF, record not found >Facility ABS: ABS_EXEC_ENVIR_NOT_FOUND, Execution env not found > SQL error code = 100, Line = 2092, File = RESD$:[SRC]EXECUTION_ENV_READ.C;1 The execution environment used by the save request is not found. Have them do a "$ABS SHOW ENVIRONMENT foo" using the execution environment name referenced in the save request. If the environment is not there you can run ABS$SYSTEM:EPCOT_DB_INIT.EXE to get the ABS default storage class/ execution environment pairs initialized. >COORDINATOR: Failed to access Storage Class >COORDINATOR: Facility ABS: ABS_ARCHIVE_ACCESS_FAILED, Failed to access Storage > Class >COORDINATOR: Line = 736, File = RESD$:[SRC]COORDINATOR.C;1 >COORDINATOR: Facility ABS: ABS_ARCHIVE_ACCESS_FAILED, Failed to access Storage > Class >COORDINATOR: Line = 370,Press RETURN to continue_ARCHIVE_MANAGEMENT.C;1 >COORDINATOR: Facility ABS: ABS_ARCHIVE_ACCESS_FAILED, Failed to access Storage > Class >COORDINATOR: Line = 704, File = RESD$:[SRC]COORD_ARCHIVE_MANAGEMENT.C;1 >COORDINATOR: Facility ABS: ABS_ARCHIVE_ACCESS_FAILED, Failed to access Storage > Class >COORDINATOR: Line = 469, File = RESD$:[SRC]ARCHIVE_FILE_SYSTEM.C;1 >COORDINATOR: Facility ABS: ABS_SLS_ALLOC_FAILED, Failed to allocate a volume v >ia SLS >COORDINATOR: Line = 1556, File = RESD$:[SRC]SLS_SERVICES.C;1 >COORDINATOR: Facility ABS: ABS_SLS_ALLOC_FAILED, Failed to allocate a volume v >ia SLS >COORDINATOR: Line = 3421, File = RESD$:[SRC]SLS_SERVICES.C;1 >COORDINATOR: Facility ABS: ABS_SLS_ALLOC_FAILED, Failed to allocate a volume v >ia SLS >COORDINATOR: Line = 3610, File = RESD$:[SRC]SLS_SERVICES.C;1 >COORDINATOR: Facility ABS: ABS_PLATFORM_SPECIFIC_ERROR, Platform-specific erro >r in diag block >COORDINATOR: %SLS-W-NETCONERR, failed to connect to network task !AS at !%D >COORDINATOR: Final status is Failed to access Storage Class ABS tried to connect to SLS, but was not able to. The SLS$DBX object was not available. Either SLS is not up and running or it is not installed/configured properly. > The GUI save request dumps him out completely - the Lui >returns the error I reported earlier. Is the customer able to use DCL to create save requests? The catalog configuration discussed in a previous note will not effect the GUI. If the catalog configuration was a problem, you would find errors in the save request log file. Is the customer upgrading from V2.0? If they are *not* upgrading from v2.0, (in other words, a new installation), the catalog conversion is not an issue. Also, please provide the entire save request log file for the failing save request. Kim | |||||
398.8 | Any updates ? | COMICS::MILLSS | "Jump! Jump now!" ...Kosh | Thu Apr 17 1997 09:43 | 38 |
Hi, I've just had a customer log a call with the same stack dump but under different circumstances. In his own words - >We have a DEC 3000-300 running VMS 7.1 and UCX 4.1 ECO 2 >We've got some temporary PAKS for testing ABS v2.1 - what we want to do is use >our VMS servers to back up NT systems. I've installed MDMS on to the Alpha >followed by ABS. We can fire up the GUI interface but it keeps crashing out. It >continually crashes on the 'Storage Class Retention' on the SAVE screen. I've >looked at the Bookreader documentation and printed this off, but it all seems a >bit dis-jointed. We've connected to the GUI interface via the Alpha console, a >VXT and Exceed under Windows 3.1 - all with similar results. The NT client side >has been installed on an NT 4 Workstation. I installed ABS from the latest CD >distribution (ie. Mar 97). >Am I missing something obvious here - we need to get this s/ware tested quickly >if it's to be considered in an NT rollout backup strategy. For completeness, I asked him for some more supporting information - >as soon as I click on the retention period option, the >ABS windows all disappear. The following are returned to >the calling DECterm procedure : > >PGRV17::SYSCON>abs/inter=decw >Exception Module Line >Exception Reraised by: PABUI_XWINDOW 848 >Exception Reraised by: PABUI_MAINWINDOW 5028 >exception PABUI_MAINWINDOW 5137 As you can see, this matches the info given in the base note. So, any ideas ? Many thanks, Simon R. Mills OpenVMS Group, UK CSC | |||||
398.8 | moved a reply | COOKIE::MHUA | Fri Apr 18 1997 10:11 | 5 | |
There was another reply posted here from COMICS::MILLSS. It's now moved to a new entry 437. Masami |