Re: Possible ZFS patch, please test!
Jeremy Chadwick wrote: On Sat, Aug 30, 2008 at 09:28:36PM +0200, Henri Hennebert wrote: John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch Just a follow-up I cvsup at Sat Aug 30 12:55 without zfs_7.patch and make buildworld make buildkernel make installkernel reboot (-s) --root on zfs is ok -- make installworld reboot System is still sluggish even during the make installworld in single user. Sorry if I've missed this, but what tuning have you done for ZFS? Some of us (most of us?) have seen fairly sluggish performance when prefetch is enabled (the default), while the system is generally more responsive when prefetch is disabled. prefetch is enabled: vfs.zfs.arc_min: 33554432 vfs.zfs.arc_max: 268435456 vfs.zfs.mdcomp_disable: 0 vfs.zfs.prefetch_disable: 0 vfs.zfs.zio.taskq_threads: 0 vfs.zfs.recover: 0 vfs.zfs.vdev.cache.size: 10485760 vfs.zfs.vdev.cache.max: 16384 vfs.zfs.cache_flush_disable: 0 vfs.zfs.zil_disable: 0 vfs.zfs.debug: 0 This may have nothing to do with the problem you've stated, but I thought I'd throw it out there. I think that the problem is somewhere else because 80% of system cpu on a dual core seems awfully bad - eg more than 20 seconds to open this response after clicking on the response button. Henri ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
John Baldwin wrote: On Friday 29 August 2008 03:57:46 am Henri Hennebert wrote: Henri Hennebert wrote: John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch It works for me with the root on zfs While rebuilding the ports index with `portsdb -Uu` the system become really sluggish with cpu running more than 60% in system... Something really strange here. Can you try removing the 'KDTRACE_*' options from your kernel config file? It appears that they haven't been enabled in 8.x by default yet. I try, but with the wold of 7.1-PRERELEASE I got cc: Internal error: Segmentation fault: 11 (program ld) Please submit a full bug report. See URL:http://gcc.gnu.org/bugs.html for instructions. *** Error code 1 Stop in /usr/obj/usr/src/sys/MORZINE. *** Error code 1 Stop in /usr/src. *** Error code 1 Stop in /usr/src. So I get /usr/bin/cc /usr/bin/ld /usr/libexec/cc* from a previous 7.0-STABLE and try it again... with a lot of cpu-system... 80% on a dual core. Anyway - I reboot the new kernel without KDTRACE_HOOKS and DDB_CTF. After reboot, system cpu is always very high and system not responsive. Henri ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
On Sat, Aug 30, 2008 at 09:07:15AM +0200, Henri Hennebert wrote: I try, but with the wold of 7.1-PRERELEASE I got cc: Internal error: Segmentation fault: 11 (program ld) Please submit a full bug report. See URL:http://gcc.gnu.org/bugs.html for instructions. *** Error code 1 Henri, please delete the entire contents of your obj directory to remove the bad tools that have been built there. When you run 'make buildkernel' it will use the tools from the last buildworld rather than the installed ones. For anyone experiencing this problem, you can do a 'make installworld' with STRIP= as long as you can boot to single user and mount your file systems. The problem is occurring when static binaries are installed with the default option to strip the binaries. It seems that the strip program doesn't like the presence of the CTF ELF section. I believe that the buildworld that you have is OK, even when built with the CTF data it's the installworld when things go bad. Do you need me to send you any files to recover from this problem? -- John Birrell ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
John Birrell wrote: On Sat, Aug 30, 2008 at 09:07:15AM +0200, Henri Hennebert wrote: I try, but with the wold of 7.1-PRERELEASE I got cc: Internal error: Segmentation fault: 11 (program ld) Please submit a full bug report. See URL:http://gcc.gnu.org/bugs.html for instructions. *** Error code 1 Henri, please delete the entire contents of your obj directory to remove the bad tools that have been built there. When you run 'make buildkernel' it will use the tools from the last buildworld rather than the installed ones. For anyone experiencing this problem, you can do a 'make installworld' with STRIP= as long as you can boot to single user and mount your file systems. The problem is occurring when static binaries are installed with the default option to strip the binaries. It seems that the strip program doesn't like the presence of the CTF ELF section. OK - I better understand what's happening. I believe that the buildworld that you have is OK, even when built with the CTF data it's the installworld when things go bad. Do you need me to send you any files to recover from this problem? No problem, I have access to a previous 7.0-STABLE. Thanks Henri -- John Birrell ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
On Sat, Aug 30, 2008 at 10:40:12AM +0200, Henri Hennebert wrote: I believe that the buildworld that you have is OK, even when built with the CTF data it's the installworld when things go bad. Do you need me to send you any files to recover from this problem? No problem, I have access to a previous 7.0-STABLE. I am concerned about the high CPU problem. All the hooks that are built in with KDTRACE_HOOKS are inactive until the DTrace modules are loaded. So there should be no CPU implications there. Are you using i3886 or amd64? -- John Birrell ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
John Birrell wrote: On Sat, Aug 30, 2008 at 10:40:12AM +0200, Henri Hennebert wrote: I believe that the buildworld that you have is OK, even when built with the CTF data it's the installworld when things go bad. Do you need me to send you any files to recover from this problem? No problem, I have access to a previous 7.0-STABLE. I am concerned about the high CPU problem. All the hooks that are built in with KDTRACE_HOOKS are inactive until the DTrace modules are loaded. So there should be no CPU implications there. Are you using i3886 or amd64? It is i386: CPU: Intel(R) Xeon(R) CPU5130 @ 2.00GHz (1995.01-MHz 686-class CPU) Origin = GenuineIntel Id = 0x6f6 Stepping = 6 Features=0xbfebfbffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE Features2=0x4e33dSSE3,RSVD2,MON,DS_CPL,VMX,TM2,SSSE3,CX16,xTPR,PDCM,DCA AMD Features=0x2010NX,LM AMD Features2=0x1LAHF Cores per package: 2 real memory = 2146369536 (2046 MB) avail memory = 2084503552 (1987 MB) Henri -- John Birrell ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch Just a follow-up I cvsup at Sat Aug 30 12:55 without zfs_7.patch and make buildworld make buildkernel make installkernel reboot (-s) --root on zfs is ok -- make installworld reboot System is still sluggish even during the make installworld in single user. Henri ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
On Sat, Aug 30, 2008 at 09:28:36PM +0200, Henri Hennebert wrote: John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch Just a follow-up I cvsup at Sat Aug 30 12:55 without zfs_7.patch and make buildworld make buildkernel make installkernel reboot (-s) --root on zfs is ok -- make installworld reboot System is still sluggish even during the make installworld in single user. Sorry if I've missed this, but what tuning have you done for ZFS? Some of us (most of us?) have seen fairly sluggish performance when prefetch is enabled (the default), while the system is generally more responsive when prefetch is disabled. This may have nothing to do with the problem you've stated, but I thought I'd throw it out there. -- | Jeremy Chadwickjdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch It works for me with the root on zfs Thanks Henri ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
Henri Hennebert wrote: John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch It works for me with the root on zfs While rebuilding the ports index with `portsdb -Uu` the system become really sluggish with cpu running more than 60% in system... Something really strange here. Henri Thanks Henri ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Possible ZFS patch, please test!
On Friday 29 August 2008 03:57:46 am Henri Hennebert wrote: Henri Hennebert wrote: John Baldwin wrote: This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch It works for me with the root on zfs While rebuilding the ports index with `portsdb -Uu` the system become really sluggish with cpu running more than 60% in system... Something really strange here. Can you try removing the 'KDTRACE_*' options from your kernel config file? It appears that they haven't been enabled in 8.x by default yet. -- John Baldwin ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Possible ZFS patch, please test!
This patch merges a few changes from HEAD back to 7.x. I think the endian changes specifically might solve the issue people saw with zpools created with non-dtrace kernels not being readable by dtrace kernels and vice versa. http://www.FreeBSD.org/~jhb/patches/zfs_7.patch -- John Baldwin ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]