Date: Sat, 21 Sep 2002 21:18:48 +0800 From: Ihsan Junaidi Ibrahim <ihsan_junaidi@yahoo.com.sg> To: freebsd-questions@freebsd.org Subject: Re: How do I know a disk is bad? Message-ID: <3399860291.20020921211848@yahoo.com.sg> In-Reply-To: <5135932470.20020919122234@yahoo.com.sg> References: <5135932470.20020919122234@yahoo.com.sg>
next in thread | previous in thread | raw e-mail | index | archive | help
Hello all, Thursday, September 19, 2002, 12:22:34 PM, freebsd-questions wrote: > Hello all, > Recent FreeBSD here. > How do I detect a bad sector on a disk. Are there any other utilities that > can be used to detect such error? I've used fsck before but I supposed > unless fsck comes up with a bunch or write/read errors, I can never know a > disk is good. Or when fsck does come with such messages, is the disk in an > inconsistent state? > The reason I ask this is because recently after making world, all file > operation (mv, cp etc.) from and to /home mounted at ad1s1e hung up and doing > kill -KILL would not cut it. I ran fsck a few times which required quite a > number of reboots just to make sure all is right. The mount point now works > as usual but I still a few doubts on how, all of sudden, things went well. > I would like to know how I can detect bad blocks on my disk and to > interpret fsck messages. > Thanks in advance. Well, I replaced the suspected disk and remount /home on another partition but the same problem persists. Sometimes an operation involving files from /home will always hang in the foreground. ps output indicates that the op is writing to disk (state D+). Apache too spawns many errors on "unable to spawn child" or "child process refused to die". I'm not very sure of why this thing is happening only to my /home FS not others. The only changes was that I remake world with 4.7 RC and enable quota on /home only. Either one of them or both could have made the problem as it is now. I did try kill -KILL, TERM on all the D, D+ state processes but couldn't hack them off. It would continue to deflate the RAM, apache especially, that I have until the very minimum forcing me to reboot and the cycle starts again whenever I try to write to /home. Whenever I would try to access the html on /home which contains scripts to write to the disk (PHP cookie, CGIs), the process would end in a D state. Subsequent operation on it respawns new apache processes which suffered the same fate and it will persist until all the RAM has been used out. I enable quotas on /home but I have not set any limit on either user or group. quotas.group and quotas.user remains as is the first time I ran edquota. At first I though a bad disk was the problem but it turned to be false although I'm not throwing that possibility out. I'm returning to 4.6.2 Release and see whether the problem persist. Has anyone had this kind of symptoms with 4.7 RC? -- Thank you for your time, Ihsan __________________________________________________ Do you Yahoo!? New DSL Internet Access from SBC & Yahoo! http://sbc.yahoo.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3399860291.20020921211848>