Re: Using an SMP machine to make multiple indices on the

Tom,

I should have forwarded you the ps output; here are the relevant lines:

*******************************************************************************
  294 ttyp0    S      0:00    203   108  1991  836  0.0 psql -e -c create
index v3_pscat_k_m_idx on v3_pscat(k_m) wsdb
  295 ?        R      0:27   2170  1425 17122 13252  1.4 postgres: postgres
wsdb [local] CREATE
  296 ttyp0    S      0:00    203   108  1991  836  0.0 psql -e -c create
index v3_pscat_h_m_idx on v3_pscat(h_m) wsdb
  297 ?        S      0:00    190  1425 11858 2436  0.2 postgres: postgres
wsdb [local] CREATE waiting
  300 ttyp0    R      0:00    273    55  3016 1384  0.1 ps avx
*******************************************************************************

Note the "CREATE waiting" process . . .

--Martin

Tom Lane wrote on Mon, 22 Oct 2001 23:09:26 EDT
>Martin Weinberg <weinberg@osprey.astro.umass.edu> writes:
>> On Postgresql 7.1.3, it seems that the table is locked after the
>> first "create index" is started up.  Is this right?
>
>AFAIK it's a share lock, which only prohibits modifications to the
>table, not reads (nor concurrent index builds).  Not sure how you
>expect the system to do better than that.
>
>            regards, tom lane
>
>---------------------------(end of broadcast)---------------------------
>TIP 5: Have you checked our extensive FAQ?
>
>http://www.postgresql.org/users-lounge/docs/faq.html
>