Re: Set new system identifier using pg_resetxlog - Mailing list pgsql-hackers

<font><span style="background-color:rgba(255,255,255,0)">Hi,<br /><br />I send you review comment about thie patch.<br
/><br/>I found no error/warning with compling and installation.<br />I have executed pg_resetxlog with some input
pattern.<br/><br />$ initdb -D data -E UTF8 --no-locale<br />$ pg_controldata data | grep "Database system
identifier"<br/>Database system identifier:           6028907917695471865<br /><br />--<br />$ pg_resetxlog -s -n  data
|grep "Database system identifier"<br /> Database system identifier:           6028907917695471865<br /><br />The -s
optiondoes not works<span></span> fine with -n option.<br /><br />--<br />$ pg_resetxlog
-s6028907917695471865111111111111111111111111111111111111111111111111111111data<br /> Transaction log reset<br />$
pg_controldatadata | grep "Database system identifier"<br />Database system identifier:
          18446744073709551615<br/><br />pg_resetxlog is finished successfully, but system identifier was not
changed.<br/> Also I think that checking data about number of digits is needed.<br /><br />regards<br />--<br />Sawada
Masahiko</span></font><br/><br />On Thursday, June 19, 2014, Petr Jelinek <<a
href="mailto:petr@2ndquadrant.com">petr@2ndquadrant.com</a>>wrote:<br /><blockquote class="gmail_quote"
style="margin:00 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On 18/06/14 19:26, Robert Haas wrote:<br
/><blockquoteclass="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> On Wed, Jun 18,
2014at 12:54 PM, Andres Freund <<a>andres@2ndquadrant.com</a>> wrote:<br /><blockquote class="gmail_quote"
style="margin:00 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> I don't see how the proposed ability makes it
moredangerous. It<br /> *already* has the ability to reset the timelineid. That's the case where<br /> users are much
morelikely to think about resetting it because that's<br /> much more plausible than taking a unrelated cluster and
resettingits<br /> sysid, timeline and LSN.<br /></blockquote><br /> All right, well, I've said my piece.  I don't have
anythingto add to<br /> that that isn't sheer repetition.  My vote is to hold off on this<br /> until we've talked
aboutreplication identifiers and other related<br /> topics in more depth.  But if that position doesn't garner
majority<br/> support ... so be it!<br /><br /></blockquote><br /> I am not sure I get what does this have to do with
replicationidentifiers. The patch has several use-cases, one of them has to do that you can know the future system id
beforeyou set it, which is useful for automating some things...<br /><br /> -- <br />  Petr Jelinek                  <a
href="http://www.2ndQuadrant.com/"target="_blank">http://www.2ndQuadrant.com/</a><br />  PostgreSQL Development, 24x7
Support,Training & Services<br /><br /><br /> -- <br /> Sent via pgsql-hackers mailing list
(<a>pgsql-hackers@postgresql.org</a>)<br/> To make changes to your subscription:<br /><a
href="http://www.postgresql.org/mailpref/pgsql-hackers"
target="_blank">http://www.postgresql.org/<u></u>mailpref/pgsql-hackers</a><br/></blockquote><br /><br />-- <br
/>Regards,<br/><br />-------<br />Sawada Masahiko<br /><br />