From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hans Reiser Subject: Re: rebuild fs Date: Tue, 05 Aug 2003 19:09:02 +0400 Message-ID: <3F2FC88E.2070403@namesys.com> References: <20030805133509.GI13689@namesys.com> <3F2FB3C6.6000407@namesys.com> <200308051819.33317.vitaly@namesys.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: list-help: list-unsubscribe: list-post: Errors-To: flx@namesys.com In-Reply-To: <200308051819.33317.vitaly@namesys.com> List-Id: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: Vitaly Fertman Cc: Oleg Drokin , Thorsten Mauch , "'reiserfs-list@namesys.com'" , vs Vitaly Fertman wrote: >>>in this case (IO error) reiserfsck does abort() which ends up as signal >>>number 5, and core is dumped if this is allowed. Looks pretty much like >>>segfault too. >>>Though a message is printed prior to this that we cannot read some block. >>> >>>Bye, >>> Oleg >>> >>> >>yuck. vs, complain to vitaly please. >> >> > >It does not look the same as the user gets different messages on the terminal. >With hardware problems like IO errors he gets "Aborting", although this can dump >the core file also. But what a user should not get even with the broken hardware >is "Segmentation fault" messages. And core dumping is what looks really pretty >much the same. > >As some old version of reiserfsck (3.6.3) stopped unexpectedly, Oleg suggested >to use the latest one -- 3.6.11 -- which worked ok for now. > >Regarding IO errors reiserfsck prints "Block ## cannot be read" before aborting >and the last ones suggest to check the hardware also. > >BTW, if there are some bad blocks I would advise to use dd_rescue instead of dd >as dd has some problems with bad blocks handling. > > > ok, you fixed the unfriendliness in latest version, excellent. -- Hans