Trevor Marshall
2002-Sep-09 01:04 UTC
Hardware - no way -Re: Crashes maybe due to: "error writing 4 unbuffered bytes"
Dear JW, I spent hours checking and upgrading my hardware. There is nothing wrong with it. It is a 'mature' Pentium-II 400MHz system that has been running my servers WITHOUT ANY CRASHES for nearly 2 years now 24/7 Listen to me: "The crashes are dues to rsync" There is an IDENTICAL report from a totally different source at http://groups.google.com/groups?q=%2B%22rsync:+error+writing+4+unbuffered+by tes+-+exiting:+Broken+pipe%22&hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=aemsfq%24894% 241%40FreeBSD.csie.NCTU.edu.tw&rnum=1 Oh - except he was Running Red Hat linux - so don't go blaming SuSe. Appreciate you trying to help, but I didn't come down in the last rain-shower :-) Trevor http://www.TrevorMarshall.com>If running rsync crashes your (non-MS) server you have a >problem with the server, not rsync. On a linux system >application caused system crashes are either bad hardware or >kernel bugs. > >Unless you have tweaked out SuSE i would say you have a >hardware problem. Rsync will stress some parts of the >system so that is probably why you haven't had crashes >before. > >Look in the system logs. The pipe errors seem more likely >to be cause by bad memory than disk. > >-- >________________________________________________________________ > J.W. Schultz Pegasystems Technologies > email address: jw@pegasys.ws > > Remember Cernan and Schmitt >
jw schultz
2002-Sep-09 01:14 UTC
Hardware - no way -Re: Crashes maybe due to: "error writing 4 unbuffered bytes"
On Sun, Sep 08, 2002 at 06:03:05PM -0700, Trevor Marshall wrote:> Dear JW, > I spent hours checking and upgrading my hardware. There is nothing wrong > with it. It is a 'mature' Pentium-II 400MHz system that has been running my > servers WITHOUT ANY CRASHES for nearly 2 years now 24/7 > > Listen to me: "The crashes are dues to rsync" > > There is an IDENTICAL report from a totally different source at > > http://groups.google.com/groups?q=%2B%22rsync:+error+writing+4+unbuffered+by > tes+-+exiting:+Broken+pipe%22&hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=aemsfq%24894% > 241%40FreeBSD.csie.NCTU.edu.tw&rnum=1 > > Oh - except he was Running Red Hat linux - so don't go blaming SuSe. > > Appreciate you trying to help, but I didn't come down in the last > rain-shower :-) >Repeating myself here, Trevor sent separately to me and here. If it isn't hardware report the kernel crashes to linux-kernel@vger.kernel.org with ksymoops et al. See kernel-newbies.org (url from memory) for instructions. When a user app crashes the kernel it is the kernel's fault.> > >If running rsync crashes your (non-MS) server you have a > >problem with the server, not rsync. On a linux system > >application caused system crashes are either bad hardware or > >kernel bugs. > > > >Unless you have tweaked out SuSE i would say you have a > >hardware problem. Rsync will stress some parts of the > >system so that is probably why you haven't had crashes > >before. > > > >Look in the system logs. The pipe errors seem more likely > >to be cause by bad memory than disk. > > > >-- > >________________________________________________________________ > > J.W. Schultz Pegasystems Technologies > > email address: jw@pegasys.ws > > > > Remember Cernan and Schmitt > > > -- > To unsubscribe or change options: http://lists.samba.org/mailman/listinfo/rsync > Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html-- ________________________________________________________________ J.W. Schultz Pegasystems Technologies email address: jw@pegasys.ws Remember Cernan and Schmitt
Jw, Respectfully, My experience is to shy away from any piece of software which the developers feel is inviolate Although often the hardware and kernel certainly could be at fault it is wrong to assume they are, and such a response usually indicates that I am wasting my time reporting this problem. Sincerely, Trevor http://TrevorMarshall.com> >If it isn't hardware report the kernel crashes to >linux-kernel@vger.kernel.org with ksymoops et al. >See kernel-newbies.org (url from memory) for instructions. > >When a user app crashes the kernel it is the kernel's fault. > > >> >> >If running rsync crashes your (non-MS) server you have a >> >problem with the server, not rsync. On a linux system >> >application caused system crashes are either bad hardware or >> >kernel bugs. >> > >> >Unless you have tweaked out SuSE i would say you have a >> >hardware problem. Rsync will stress some parts of the >> >system so that is probably why you haven't had crashes >> >before. >> > >> >Look in the system logs. The pipe errors seem more likely >> >to be cause by bad memory than disk. >> > >> >-- >> >________________________________________________________________ >> > J.W. Schultz Pegasystems Technologies >> > email address: jw@pegasys.ws >> > >> > Remember Cernan and Schmitt >> > >> -- >> To unsubscribe or change options:http://lists.samba.org/mailman/listinfo/rsync>> Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html > >-- >________________________________________________________________ > J.W. Schultz Pegasystems Technologies > email address: jw@pegasys.ws > > Remember Cernan and Schmitt >-- >To unsubscribe or change options:http://lists.samba.org/mailman/listinfo/rsync>Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html >
I agree often if the kernel crashes, though the kernel should be well protected enough to only crash the application process, and not damage itself, often the kernel flaw is stimulated by an errant or out of range value in it's input which woudl cause the user process to crash if the kernel did not. That said, to clarify to them. Did the whole server machine crash ie: the kernel, or just the rsync process or login process it was invoked from? PK At 06:29 PM 9/8/2002 -0700, you wrote:>Jw, >Respectfully, >My experience is to shy away from any piece of software which the >developers feel is inviolate > >Although often the hardware and kernel certainly could be at fault it is >wrong to assume they are, and such a response usually indicates that I am >wasting my time reporting this problem. > >Sincerely, >Trevor >http://TrevorMarshall.com > > > > >If it isn't hardware report the kernel crashes to > >linux-kernel@vger.kernel.org with ksymoops et al. > >See kernel-newbies.org (url from memory) for instructions. > > > >When a user app crashes the kernel it is the kernel's fault. > > > > > >> > >> >If running rsync crashes your (non-MS) server you have a > >> >problem with the server, not rsync. On a linux system > >> >application caused system crashes are either bad hardware or > >> >kernel bugs. > >> > > >> >Unless you have tweaked out SuSE i would say you have a > >> >hardware problem. Rsync will stress some parts of the > >> >system so that is probably why you haven't had crashes > >> >before. > >> > > >> >Look in the system logs. The pipe errors seem more likely > >> >to be cause by bad memory than disk. > >> > > >> >-- > >> >________________________________________________________________ > >> > J.W. Schultz Pegasystems Technologies > >> > email address: jw@pegasys.ws > >> > > >> > Remember Cernan and Schmitt > >> > > >> -- > >> To unsubscribe or change options: >http://lists.samba.org/mailman/listinfo/rsync > >> Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html > > > >-- > >________________________________________________________________ > > J.W. Schultz Pegasystems Technologies > > email address: jw@pegasys.ws > > > > Remember Cernan and Schmitt > >-- > >To unsubscribe or change options: >http://lists.samba.org/mailman/listinfo/rsync > >Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html > > >-- >To unsubscribe or change options: >http://lists.samba.org/mailman/listinfo/rsync >Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html
At 09:36 PM 9/8/2002 -0400, you wrote:>I agree often if the kernel crashes, though the kernel should be well >protected enough to only crash the application process, and not damage >itself, often the kernel flaw is stimulated by an errant or out of range >value in it's input which woudl cause the user process to crash if the >kernel did not. > >That said, to clarify to them. Did the whole server machine crash ie: the >kernel, or just the rsync process or login process it was invoked from? >PKThanks, Peter, I had 2 SSH threads running to the machine (on eth0 192.168.1.0/24), they were being used to issue commands, look at logs and review progress. I was usually not logged in at console. When rsync/kernel/hardware crashed I would find that my remote SSH consoles would go dead and all online http tasks would not respond either (on eth1). The main console was still at the login: prompt, and if I entered a name and hit return, it would echo the name and return, but never get to the password stage. What was interesting is that as I incrementally rebooted, fsck'd and did succesive snapshots, I eventually ended up with a complete 2gig copy of my root filesystem, and rsync exited normally from that run. But the next run it crashed on a 680 megabyte logfile, which I renamed so it was no longer going to be open. Next run it gave me the reported error message on a smaller, but also presumably open, logfile. Here is the command from my scipt: $RSYNC \ -va --delete \ --exclude="root/snapshot" \ --exclude="proc/" \ --exclude="dev/" \ /. $SNAPSHOT/daily-0 ; Lets see, oh... /proc/version "Linux version 2.4.10-4GB (root@Pentium.suse.de) (gcc version 2.95.3 20010315 (SuSE)) #1" Trevor