I find that my killed connection server process process disappear in Status
after a long time.
> If pgAdmin's cancel is just dropping the connection, the server might
> take some time to notice it, especially if it's in the process of running
> a query and doesn't have reason to talk to pgAdmi
On Mar 22, 2006, at 7:14 PM, Andrus wrote:
Do you mean that the statement hadn't been fully transmitted yet?
Yes. Sending 2 MB takes most of time so client is terminated in
this stage.
If so, the backend would have just been waiting for the rest of the
statement to arrive. Perhaps you'r
> Do you mean that the statement hadn't been fully transmitted yet?
Yes. Sending 2 MB takes most of time so client is terminated in this stage.
> If so, the backend would have just been waiting for the rest of the
> statement to arrive. Perhaps you're looking at some sort of network bug
> that c
"Andrus" <[EMAIL PROTECTED]> writes:
>> What's it doing exactly?
> odbc driver was in progress of sending data using libpq when client was
> terminated.
Do you mean that the statement hadn't been fully transmitted yet?
If so, the backend would have just been waiting for the rest of the
statement
>> A cancel signal was sent to the selected server process(es).
>> However, process continues running.
>
> What's it doing exactly?
Client was killed when it waits a 2 MB bytea string insert statement to be
completed inside transaction:
INSERT INTO localfil (filename,filedirect,BlockNumber,lastc
"Andrus" <[EMAIL PROTECTED]> writes:
> I tried to kill the process using pgAdmin Status page Cancel button.
> pgAdmin responds:
> A cancel signal was sent to the selected server process(es).
> However, process continues running.
What's it doing exactly?
We recently fixed some places in btree inde
>> It is not because of the locks. There is only running transaction (and it
>> got every locks). Not sure why the backend stucks there, seems it doesn't
>> aware of the broken client. In normal situations, if you killed a client,
>> then the server will print something like "could not receive data
"Andrus" <[EMAIL PROTECTED]> wrote
>
> insert command takes a long time. I terminated the client application from
> Windows XP client manager during insert command.
> At next re-run application and pgadmin both hang in DELETE FROM line
>
> I think I can reproduce this. Postgres server is in FreeBS
> It is not because of the locks. There is only running transaction (and it
> got every locks). Not sure why the backend stucks there, seems it doesn't
> aware of the broken client. In normal situations, if you killed a client,
> then the server will print something like "could not receive data fro
"Andrus" <[EMAIL PROTECTED]> wrote
> After killing client from XP task manager command
>
> DELETE FROM mytable
>
> runs forever.
>
> This is probably because pgAdmin shows that there are some locks placed
(see
> included picture).
>
It is not because of the locks. There is only running transactio
10 matches
Mail list logo
