Обсуждение: xlog related error
Hi, A friend of mine getting following error Error: xlog flush request 0/797FED50/ is not satisfied -- flushed only to 0/797FC1D8 CONTEXT: writing block 681 of relation 1663/10819/16520 Detail: Multiple failures -- write error may be permanent can anyone pls suggest if its indicative of some certain situation. regds mallah.
"Rajesh Kumar Mallah" <mallah.rajesh@gmail.com> writes:
> A friend of mine getting following error
> Error: xlog flush request 0/797FED50/ is not satisfied -- flushed only
> to 0/797FC1D8
> CONTEXT: writing block 681 of relation 1663/10819/16520
Which PG version exactly? Has he had any crashes lately? The message
means that that block contains an LSN that's past the current end of WAL,
which is a pretty suspicious situation.
The message will go away pretty soon (as soon as another 10K or so of
WAL is generated) but this looks like a symptom of a bug to me. Details
would be good.
regards, tom lane
On 4/5/07, Tom Lane <tgl@sss.pgh.pa.us> wrote: > "Rajesh Kumar Mallah" <mallah.rajesh@gmail.com> writes: > > A friend of mine getting following error > > > Error: xlog flush request 0/797FED50/ is not satisfied -- flushed only > > to 0/797FC1D8 > > CONTEXT: writing block 681 of relation 1663/10819/16520 > > Which PG version exactly? Has he had any crashes lately? The message > means that that block contains an LSN that's past the current end of WAL, > which is a pretty suspicious situation. > > The message will go away pretty soon (as soon as another 10K or so of > WAL is generated) but this looks like a symptom of a bug to me. Details > would be good. The person "Ashwin" might contact the list directly. If he does'nt i shall try my best to get the details regds mallah. > > regards, tom lane >
--- Rajesh Kumar Mallah <mallah.rajesh@gmail.com> wrote: > On 4/5/07, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > "Rajesh Kumar Mallah" <mallah.rajesh@gmail.com> > writes: > > > A friend of mine getting following error > > > > > Error: xlog flush request 0/797FED50/ is not > satisfied -- flushed only > > > to 0/797FC1D8 > > > CONTEXT: writing block 681 of relation > 1663/10819/16520 > > > > Which PG version exactly? Has he had any crashes > lately? The message > > means that that block contains an LSN that's past > the current end of WAL, > > which is a pretty suspicious situation. > > > > The message will go away pretty soon (as soon as > another 10K or so of > > WAL is generated) but this looks like a symptom of > a bug to me. Details > > would be good. > > The person "Ashwin" might contact the list directly. > If he does'nt i shall try my best to get the details > > regds mallah. > > > > > regards, tom lane > > > > ---------------------------(end of > broadcast)--------------------------- > TIP 2: Don't 'kill -9' the postmaster > I am using postgres-8.2.3, i have changed my postgres.c file, in the function PostgresMain in the case P of processing the command so that, whenever i see a word say "tiger" as the first word of the query, i will generate 10,000 inserts into a table and execute them using exec_simple. I will fire one query through jdbc beginning with the same word "tiger". Here after inserting 5600 inserts i get the above ERROR message. Any idea how to succesfully insert all those values. Regards, Ashwin __________________________________________________________ Yahoo! India Answers: Share what you know. Learn something new http://in.answers.yahoo.com/
>Rajesh Kumar Mallah <mallah ( dot ) rajesh ( at ) gmail ( dot ) com>
>wrote:
>> On 4/5/07, Tom Lane <tgl ( at ) sss ( dot ) pgh ( dot ) pa ( dot ) us> wrote:
>> > "Rajesh Kumar Mallah" <mallah ( dot ) rajesh ( at ) gmail ( dot ) com>
>> writes:
>> > > A friend of mine getting following error
>> >
>> > > Error: xlog flush request 0/797FED50/ is not
>> satisfied -- flushed only
>> > > to 0/797FC1D8
>> > > CONTEXT: writing block 681 of relation
>> 1663/10819/16520
>> >
>> > Which PG version exactly? Has he had any crashes
>> lately? The message
>> > means that thatblock contains an LSN that's past
>> the current end of WAL,
>> > which is a pretty suspicious situation.
>> >
>> > The message will go away pretty soon (as soon as
>> another 10K or so of
>> > WAL is generated) but this looks like a symptom of
>> a bug to me. Details
> > would be good.
> >
>> The person "Ashwin" might contact the list directly.
>> If he does'nt i shall try my best to get the details
>>
>> regds mallah.
>>
>> >
>> > regards, tom lane
>> >
>>
>> ---------------------------(end of
>> broadcast)---------------------------
>> TIP 2: Don't 'kill -9' the postmaster
>>
>I am using postgres-8.2.3, i have changed my
>postgres.c file, in the function PostgresMain in the
>case P of processing the commandso that, whenever i
>see a word say "tiger" as the first word of the
>query, i will generate 10,000 inserts into a table and
>execute them using exec_simple.
>I will fire one query through jdbc beginning with the
>same word "tiger".
>Here after inserting 5600 inserts i get the above
>ERROR message.
>Any idea how to succesfully insert all those values.
>Regards,
>Ashwin
I have done pg_restore as i found in some website about this error.
Now i am not getting the error but now my postgres stops after inserting
5600(+/- 50) rows. I cant find out where the problem is
Regards,
Ashwin
Here’s a new way to find what you're looking for - Yahoo! Answers
Bellur Ashwin <bell_ashwin@yahoo.com> writes:
>> I am using postgres-8.2.3, i have changed my
>> postgres.c file, in the function PostgresMain in the
>> case P of processing the command so that, whenever i
>> see a word say "tiger" as the first word of the
>> query, i will generate 10,000 inserts into a table and
>> execute them using exec_simple.
>> I will fire one query through jdbc beginning with the
>> same word "tiger".
>> Here after inserting 5600 inserts i get the above
>> ERROR message.
>> Any idea how to succesfully insert all those values.
The odds seem extremely high that you broke something while changing
postgres.c. pgsql-admin is hardly the place to ask about hacking
backend innards...
regards, tom lane