<HTML><BODY><br><br><br><blockquote style="border-left:1px solid #0857A6; margin:10px; padding:0 0 0 10px;">
Четверг, 13 февраля 2020, 9:48 +03:00 от Konstantin Osipov <kostja.osipov@gmail.com>:<br><br><div id=""><div class="js-helper js-readmsg-msg"><div><div id="style_15815765130937420987_BODY">* sergepetrenko <<a href="mailto:sergepetrenko@tarantool.org">sergepetrenko@tarantool.org</a>> [20/02/13 09:34]:<br><div class="mail-quote-collapse">> There is an assertion in vclock_follow `lsn > prev_lsn`, which doesn't<br>
> fire in release builds, of course. So we better panic on an attemt to<br>
> write a record with a duplicate or otherwise broken lsn.<br>
> <br></div><br>
It should be a warning in production. You can't crash production<br>
deploy because of a server bug, and lsn inconsistency is always a<br>
server bug.</div></div></div></div></blockquote><br>Well, this inconsistency yields a WAL that cannot be recovered from.<br>But I don't have a strong opinion here. I changed it to a warning in v2.<br><blockquote style="border-left:1px solid #0857A6; margin:10px; padding:0 0 0 10px;"><div id=""><div class="js-helper js-readmsg-msg"><div><div id="style_15815765130937420987_BODY"><br><br>
-- <br>
Konstantin Osipov, Moscow, Russia<br></div></div></div></div></blockquote>
<br>
<br>-- <br>Sergey Petrenko<br></BODY></HTML>