Thread: BUG #17867: pg_dump intermittently hangs
The following bug has been logged on the website: Bug reference: 17867 Logged by: Mayur Lokare Email address: mayur.lokare%veritas.com@gtempaccount.com PostgreSQL version: 14.6 Operating system: Red Hat Universal Base Image 8/ UBI8 Description: Database demo contains 667 tables and around 30 GB of data. pg_dump command : pg_dump host=server.postgres.database.azure.com -p 5432 username=admin1@server -w -F c -dbname=demo -v -f /root/out.dump Getting below stack: (gdb) bt #0 0x00007f6ab8ea6bc8 in write () from /lib64/libc.so.6 #1 0x00007f6ab8e18dad in _IO_file_write@@GLIBC_2.2.5 () from /lib64/libc.so.6 #2 0x00007f6ab8e1811f in new_do_write () from /lib64/libc.so.6 #3 0x00007f6ab8e194de in _GI_IO_file_xsputn () from /lib64/libc.so.6 #4 0x00007f6ab8e0e56c in fwrite () from /lib64/libc.so.6 #5 0x000000000043d861 in flushbuffer () #6 0x000000000043f064 in pg_vfprintf () #7 0x000000000043f148 in pg_fprintf () #8 0x000000000043bf4a in pg_log_generic_v () #9 0x000000000043c1d8 in pg_log_generic () #10 0x00000000004090d9 in dumpTableData_copy () #11 0x000000000043195b in WriteDataChunksForTocEntry () #12 0x00000000004319e6 in WriteDataChunks () #13 0x0000000000433199 in _CloseArchive () #14 0x000000000042c92e in CloseArchive () #15 0x0000000000406536 in main () on another database userdb, we have 5193 tables and around 60 mb of data pg_dump host=server.postgres.database.azure.com -p 5432 username=admin1@server -w -F c -dbname=userdb -v -f /root/out2.dump (gdb) bt #0 0x00007fe6a9920bc8 in write () from /lib64/libc.so.6 #1 0x00007fe6a9892dad in _IO_file_write@@GLIBC_2.2.5 () from /lib64/libc.so.6 #2 0x00007fe6a989211f in new_do_write () from /lib64/libc.so.6 #3 0x00007fe6a98934de in __GI__IO_file_xsputn () from /lib64/libc.so.6 #4 0x00007fe6a988856c in fwrite () from /lib64/libc.so.6 #5 0x000000000043d861 in flushbuffer () #6 0x000000000043f064 in pg_vfprintf () #7 0x000000000043f148 in pg_fprintf () #8 0x000000000043bf4a in pg_log_generic_v () #9 0x000000000043c1d8 in pg_log_generic () #10 0x0000000000421db1 in getTableAttrs () #11 0x0000000000425431 in getSchemaData () #12 0x00000000004059a1 in main ()
> On 24 Mar 2023, at 09:02, PG Bug reporting form <noreply@postgresql.org> wrote: > > The following bug has been logged on the website: > > Bug reference: 17867 > Logged by: Mayur Lokare > Email address: mayur.lokare%veritas.com@gtempaccount.com > PostgreSQL version: 14.6 > Operating system: Red Hat Universal Base Image 8/ UBI8 > Description: > > Database demo contains 667 tables and around 30 GB of data. > > pg_dump command : > pg_dump host=server.postgres.database.azure.com -p 5432 > username=admin1@server -w -F c -dbname=demo -v -f /root/out.dump > > Getting below stack: > (gdb) bt > #0 0x00007f6ab8ea6bc8 in write () from /lib64/libc.so.6 > #1 0x00007f6ab8e18dad in _IO_file_write@@GLIBC_2.2.5 () from > /lib64/libc.so.6 > #2 0x00007f6ab8e1811f in new_do_write () from /lib64/libc.so.6 > #3 0x00007f6ab8e194de in _GI_IO_file_xsputn () from /lib64/libc.so.6 > #4 0x00007f6ab8e0e56c in fwrite () from /lib64/libc.so.6 > #5 0x000000000043d861 in flushbuffer () Is is really hanging and not just waiting on IO? -- Daniel Gustafsson
PG Bug reporting form <noreply@postgresql.org> writes: > pg_dump command : > pg_dump host=server.postgres.database.azure.com -p 5432 > username=admin1@server -w -F c -dbname=demo -v -f /root/out.dump Both of the stack traces you show are blocked trying to write debug messages to the terminal. Maybe you should drop "-v". regards, tom lane