Thread: Re: sudden program termination: no warning, error, or crash
I am also getting repeated sudden program terminations. My environment is Windows Server 2003 Native Japanese OS. Postgresql8.4. Pgadmin III 1.10. Often after just having added a column, the program terminates. In all cases thus far, the operation that I was performing(addition to or maintenance of a column) was successful (i.e., the change was saved). -----Original Message----- From: pgadmin-support-owner@postgresql.org [mailto:pgadmin-support-owner@postgresql.org] On Behalf Of Kevin Field Sent: Wednesday, August 12, 2009 12:06 AM To: pgadmin-support@postgresql.org Subject: [pgadmin-support] sudden program termination: no warning, error, or crash This is a continuation of http://groups.google.com/group/pgsql.interfaces.pgadmin.support/browse_thread/thread/4b919debf3362670/ee431c94c2c9299?q=#0ee431c94c2c9299 (just after a certain amount of time Google groups doesn't let you reply to a thread anymore.) -------------------- So on the latest distribution, I had this happen again. This time, I was clicking on a view node in the tree of an 8.4 server. The program cleared the SQL pane and paused for a moment to query the server, as it normally does, and then all of the sudden pgAdmin and a query editor window I had opened and had been sitting idle disappeared from the screen. In the Event Viewer I found this. Event Type: Error Event Source: Application Error Event Category: (100) Event ID: 1000 Date: 2009/08/11 Time: 10:57:46 AM User: N/A Computer: OUT Description: Faulting application pgAdmin3.exe, version 1.10.0.6, faulting module pgAdmin3.exe, version 1.10.0.6, fault address 0x001e0ca8. For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp. Data: 0000: 41 70 70 6c 69 63 61 74 Applicat 0008: 69 6f 6e 20 46 61 69 6c ion Fail 0010: 75 72 65 20 20 70 67 41 ure pgA 0018: 64 6d 69 6e 33 2e 65 78 dmin3.ex 0020: 65 20 31 2e 31 30 2e 30 e 1.10.0 0028: 2e 36 20 69 6e 20 70 67 .6 in pg 0030: 41 64 6d 69 6e 33 2e 65 Admin3.e 0038: 78 65 20 31 2e 31 30 2e xe 1.10. 0040: 30 2e 36 20 61 74 20 6f 0.6 at o 0048: 66 66 73 65 74 20 30 30 ffset 00 0050: 31 65 30 63 61 38 1e0ca8 Kev -- Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgadmin-support =
Hello, Thursday, August 13, 2009, 2:38:02 AM, you wrote: JM> I am also getting repeated sudden program terminations. My JM> environment is Windows Server 2003 Native Japanese OS. Postgresql JM> 8.4. Pgadmin III 1.10. Often after just having added a column, the JM> program terminates. In all cases thus far, the operation that I JM> was performing (addition to or maintenance of a column) was JM> successful (i.e., the change was saved). Although this won't be helpful, but it happened to me as well yesterday. WinXPSP3 pgAdmin1.10 connected to a PostgreSQL 8.4 database. I was left clicking on group/user roles in the tree view when pgAdmin crashed and the window disappeared. After restarting the program I tried to reproduce this behaviour, but the crash was not reproducable. Rainer
When I'm doing heavy development I'm always removing and recreating tables and views. If you - drop a view - create a view with the same name - and have the data window for the view open from before it was dropped - and do a refresh You lose the column names, since the oid for the view/table is cached. So if upon refresh, it rechecked the oid of the thing you are viewing, that would fix the issue.
If you - have a query window open - and you're editing a function definition, or trigger - execute the query, and update/replace the function or trigger, or type. or table, or any other object - and then later right click on the object and do 'create script' you'll get the old data that was cached from the last refresh I can't count how many times I've lost changes because I made new changes from old data using 'create script' If a refresh is done, to update the data, and then the create script window is opened, that would be awesome.
Thanks - logged as a todo: http://code.pgadmin.org/trac/ticket/43 On Fri, Aug 14, 2009 at 3:24 AM, Mark Murawski<kobaz@kobaz.net> wrote: > If you > - have a query window open > - and you're editing a function definition, or trigger > - execute the query, and update/replace the function or trigger, or type. or > table, or any other object > - and then later right click on the object and do 'create script' > > you'll get the old data that was cached from the last refresh > I can't count how many times I've lost changes because I made new changes > from old data using 'create script' > > If a refresh is done, to update the data, and then the create script window > is opened, that would be awesome. > > > > -- > Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgadmin-support > -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
Thanks, logged: http://code.pgadmin.org/trac/ticket/44 On Fri, Aug 14, 2009 at 3:20 AM, Mark Murawski<kobaz@kobaz.net> wrote: > When I'm doing heavy development I'm always removing and recreating tables > and views. > > If you > - drop a view > - create a view with the same name > - and have the data window for the view open from before it was dropped > - and do a refresh > > You lose the column names, since the oid for the view/table is cached. So > if upon refresh, it rechecked the oid of the thing you are viewing, that > would fix the issue. > > > -- > Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgadmin-support > -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
On Thu, Aug 13, 2009 at 8:45 AM, Rainer Bauer<usenet@munnin.com> wrote: > Hello, > > Thursday, August 13, 2009, 2:38:02 AM, you wrote: > > JM> I am also getting repeated sudden program terminations. My > JM> environment is Windows Server 2003 Native Japanese OS. Postgresql > JM> 8.4. Pgadmin III 1.10. Often after just having added a column, the > JM> program terminates. In all cases thus far, the operation that I > JM> was performing (addition to or maintenance of a column) was > JM> successful (i.e., the change was saved). > > Although this won't be helpful, but it happened to me as well > yesterday. WinXPSP3 pgAdmin1.10 connected to a PostgreSQL 8.4 > database. > > I was left clicking on group/user roles in the tree view when pgAdmin > crashed and the window disappeared. > > After restarting the program I tried to reproduce this behaviour, but > the crash was not reproducable. I'm not sure how we'll be able to track down these cases without a reproducible test case. That said, I've been fixing some bugs that cause similar crashes on Solaris which seems to be *far* more picky about some things than other OSs. From the description that jwexler gave, it's not beyond the realms of possibility that you are seeing the same/similar bugs, just with far less frequency. -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
This is probably the same issue as I reported in an earlier thread "Sudden crash in pgadmin" I haven't changed anything on this machine so I can reproduce it if needed. Andrew On Mon, Aug 17, 2009 at 8:26 PM, Dave Page<dpage@pgadmin.org> wrote: > On Thu, Aug 13, 2009 at 8:45 AM, Rainer Bauer<usenet@munnin.com> wrote: >> Hello, >> >> Thursday, August 13, 2009, 2:38:02 AM, you wrote: >> >> JM> I am also getting repeated sudden program terminations. My >> JM> environment is Windows Server 2003 Native Japanese OS. Postgresql >> JM> 8.4. Pgadmin III 1.10. Often after just having added a column, the >> JM> program terminates. In all cases thus far, the operation that I >> JM> was performing (addition to or maintenance of a column) was >> JM> successful (i.e., the change was saved). >> >> Although this won't be helpful, but it happened to me as well >> yesterday. WinXPSP3 pgAdmin1.10 connected to a PostgreSQL 8.4 >> database. >> >> I was left clicking on group/user roles in the tree view when pgAdmin >> crashed and the window disappeared. >> >> After restarting the program I tried to reproduce this behaviour, but >> the crash was not reproducable. > > I'm not sure how we'll be able to track down these cases without a > reproducible test case. That said, I've been fixing some bugs that > cause similar crashes on Solaris which seems to be *far* more picky > about some things than other OSs. From the description that jwexler > gave, it's not beyond the realms of possibility that you are seeing > the same/similar bugs, just with far less frequency. > > -- > Dave Page > EnterpriseDB UK: http://www.enterprisedb.com > > -- > Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgadmin-support > -- ___________________________________________ Andrew J. P. Maclean Centre for Autonomous Systems The Rose Street Building J04 The University of Sydney 2006 NSW AUSTRALIA Ph: +61 2 9351 3283 Fax: +61 2 9351 7474 URL: http://www.acfr.usyd.edu.au/ ___________________________________________
I am re-sending this message because I received a message back (permanent error) from someone's mail server that the message was greylisted (i.e., ensuring authentic email by requiring sender to re-send message). For those who already received this, apologies for the duplicate. The message: I can of course reproduce as well. The occurrence might be associated with Shift-Tab to toggle to excel immediately after a column edit operation (i.e., while still processing the transaction). I've described in detail below the steps that most often reproduce the termination: 1. While an existing column is selected, click on the "Create a new object of the same type as the selected object." Button. 2. Toggle to an Excel 2007 approx. 450kb spreadsheet and copy "creationTimestamp" text from within a cell (i.e., only text itself is copied not the cell) 3. Toggle back to pgAdmin III and paste in "Name" 4. Set Data Type as "timestamp with time zone" 5. Click "OK" button to close window 6. Immediately after click "OK" button (i.e., very promptly and perhaps while pgAdmin III is still processing), do a Shift-Tab to toggle back to Excel 2007. 7. In Excel, copy the text "Date and time record was created". Toggle back to pgAdmin III via Shift-Tab. 8. Select "creationTimestamp" in object browser and click on "Display/edit the properties of the selected object." 9. Paste the copied text into the Comment field. 10. Click "OK" button to close window 11. Immediately after click "OK" button (i.e., very promptly and perhaps while pgAdmin III is still processing), do a Shift-Tab to toggle back to Excel 2007. Sometimes crashes at this point with a pause at the pgAdmin III screen and then the toggle to Excel after pgAdmin III has terminated. If it has not terminated then continuation with below steps. 12. Repeat steps #1-2 but this time for "updateTimestamp" text 13. Repeat steps #3-6. Often terminates at this point when doing the same thing as #6. However, less occurrence of the termination if I adequately pause before doing the Shift-Tab back to Excel. Jeff -----Original Message----- From: Andrew Maclean [mailto:andrew.amaclean@gmail.com] Sent: Wednesday, August 19, 2009 9:20 AM To: Dave Page Cc: Rainer Bauer; pgadmin-support@postgresql.org; JwexlerAt MailDotCom Subject: Re: [pgadmin-support] sudden program termination: no warning, error, or crash This is probably the same issue as I reported in an earlier thread "Sudden crash in pgadmin" I haven't changed anything on this machine so I can reproduce it if needed. Andrew On Mon, Aug 17, 2009 at 8:26 PM, Dave Page<dpage@pgadmin.org> wrote: > On Thu, Aug 13, 2009 at 8:45 AM, Rainer Bauer<usenet@munnin.com> wrote: >> Hello, >> >> Thursday, August 13, 2009, 2:38:02 AM, you wrote: >> >> JM> I am also getting repeated sudden program terminations. My >> JM> environment is Windows Server 2003 Native Japanese OS. Postgresql >> JM> 8.4. Pgadmin III 1.10. Often after just having added a column, the >> JM> program terminates. In all cases thus far, the operation that I >> JM> was performing (addition to or maintenance of a column) was >> JM> successful (i.e., the change was saved). >> >> Although this won't be helpful, but it happened to me as well >> yesterday. WinXPSP3 pgAdmin1.10 connected to a PostgreSQL 8.4 >> database. >> >> I was left clicking on group/user roles in the tree view when pgAdmin >> crashed and the window disappeared. >> >> After restarting the program I tried to reproduce this behaviour, but >> the crash was not reproducable. > > I'm not sure how we'll be able to track down these cases without a > reproducible test case. That said, I've been fixing some bugs that > cause similar crashes on Solaris which seems to be *far* more picky > about some things than other OSs. From the description that jwexler > gave, it's not beyond the realms of possibility that you are seeing > the same/similar bugs, just with far less frequency. > > -- > Dave Page > EnterpriseDB UK: http://www.enterprisedb.com > > -- > Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgadmin-support > -- ___________________________________________ Andrew J. P. Maclean Centre for Autonomous Systems The Rose Street Building J04 The University of Sydney 2006 NSW AUSTRALIA Ph: +61 2 9351 3283 Fax: +61 2 9351 7474 URL: http://www.acfr.usyd.edu.au/ ___________________________________________
On Wed, Aug 19, 2009 at 1:19 AM, Andrew Maclean<andrew.amaclean@gmail.com> wrote: > This is probably the same issue as I reported in an earlier thread > "Sudden crash in pgadmin" > I haven't changed anything on this machine so I can reproduce it if needed. Looking back over that thread, it looks like at least some of it was off-list, between you and Ashesh. Did anything come of it? -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
Please keep the thread on-list to avoid confusion like we had previously. On Wed, Aug 19, 2009 at 11:40 AM, Andrew Maclean<a.maclean@cas.edu.au> wrote: > No the log files didn't show anything unusual. If you want, I can re-run it > tomorrow and capture the log file. Please do. Thanks. -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
Sorry I was on a different machine. Here is the log file. Essentially the whole program crashes when you connect to a database. Andrew On Wed, Aug 19, 2009 at 8:42 PM, Dave Page<dpage@pgadmin.org> wrote: > Please keep the thread on-list to avoid confusion like we had previously. > > On Wed, Aug 19, 2009 at 11:40 AM, Andrew Maclean<a.maclean@cas.edu.au> wrote: >> No the log files didn't show anything unusual. If you want, I can re-run it >> tomorrow and capture the log file. > > Please do. > > Thanks. > > -- > Dave Page > EnterpriseDB UK: http://www.enterprisedb.com > > -- > Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgadmin-support > -- ___________________________________________ Andrew J. P. Maclean Centre for Autonomous Systems The Rose Street Building J04 The University of Sydney 2006 NSW AUSTRALIA Ph: +61 2 9351 3283 Fax: +61 2 9351 7474 URL: http://www.acfr.usyd.edu.au/ ___________________________________________
Attachment
On Thu, Aug 20, 2009 at 12:08 AM, Andrew Maclean<andrew.amaclean@gmail.com> wrote: > Sorry I was on a different machine. > Here is the log file. > Essentially the whole program crashes when you connect to a database. That sure ain't normal. There's nothing in the log - does it happen with just one server, or can you reproduce it with more than one? Do you see the crash on multiple machines? -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com
Unfortunately it is a remote machine and can only connect to that server. So I can't test that. On Fri, Aug 21, 2009 at 1:39 AM, Dave Page<dpage@pgadmin.org> wrote: > On Thu, Aug 20, 2009 at 12:08 AM, Andrew > Maclean<andrew.amaclean@gmail.com> wrote: >> Sorry I was on a different machine. >> Here is the log file. >> Essentially the whole program crashes when you connect to a database. > > That sure ain't normal. There's nothing in the log - does it happen > with just one server, or can you reproduce it with more than one? Do > you see the crash on multiple machines? > > -- > Dave Page > EnterpriseDB UK: http://www.enterprisedb.com > > -- > Sent via pgadmin-support mailing list (pgadmin-support@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgadmin-support > -- ___________________________________________ Andrew J. P. Maclean Centre for Autonomous Systems The Rose Street Building J04 The University of Sydney 2006 NSW AUSTRALIA Ph: +61 2 9351 3283 Fax: +61 2 9351 7474 URL: http://www.acfr.usyd.edu.au/ ___________________________________________