Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 05:26:25PM +0200, Andrea Arcangeli wrote: > You would help me if you could try to do the same with 2.2.18pre15aa1 on both Jeff, don't waste time trying it because Jay Weber just fixed the bug. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 05:26:25PM +0200, Andrea Arcangeli wrote: You would help me if you could try to do the same with 2.2.18pre15aa1 on both Jeff, don't waste time trying it because Jay Weber just fixed the bug. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 11:20:14AM -0400, Jeff Garzik wrote: > No problem here, I always use 8192 r/w size :) You would help me if you could try to do the same with 2.2.18pre15aa1 on both client and server side nfsv3 using also files larger than 2G. Maybe the problem with the bigger r/w size isn't a linux problem but maybe any kind of network problem that triggers increasing the UDP packet size. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
Andrea Arcangeli wrote: > On Fri, Oct 13, 2000 at 10:54:50AM -0400, Jeff Garzik wrote: > > I've tested 2.2.18pre15 (w/out 'aa1') against FreeBSD, Solaris, and > > other Linux clients, and I'm very happy with the improvement over older > > Linux NFS. All of my testing and real-life usage was done w/ NFSv3 > > enabled. > > Could you try to use 8192 r/w size? That doesn't happen with the aa patchkit > either without changing the packet size. No problem here, I always use 8192 r/w size :) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 10:54:50AM -0400, Jeff Garzik wrote: > Andrea Arcangeli wrote: > > > > On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: > > > Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches > > > applied and some other stuff that we needed, and the machines are stable > > > > I recommend using 2.2.18pre15aa1 (without using nfsv3 in production for now). > > Why, specifically, no NFSv3? Do you have a reason or just paranoia? :) I've a bug report pending for NFSv3 on 2.2.18pre15aa1, it's not paranoia. But I'm not really talking about 2.2.18pre15, it may be a bug in the aa patchkit. OTOH Andi couldn't reproduce it I will try soon too. > I've tested 2.2.18pre15 (w/out 'aa1') against FreeBSD, Solaris, and > other Linux clients, and I'm very happy with the improvement over older > Linux NFS. All of my testing and real-life usage was done w/ NFSv3 > enabled. Could you try to use 8192 r/w size? That doesn't happen with the aa patchkit either without changing the packet size. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 04:23:51PM +0200, David Weinehall wrote: > On Fri, Oct 13, 2000 at 04:14:32PM +0200, Andrea Arcangeli wrote: > > On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: > > > Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli > > > patches applied and some other stuff that we needed, and the > > > machines are stable > > > > I recommend using 2.2.18pre15aa1 (without using nfsv3 in production > > for now). > > I'd say that NOT using NFSv3 for production is nothing but folly. > The NFSv3 that's in v2.2.18pre15 seems to work VERY fine (we've beaten I were talking only about 2.2.18pre15aa1. Sorry if I wasn't clear. v2.2.18pre15 has nfsv3 but it's not LFS capable. (however 2.2.18pre15 _may_ have the same problems that I got reported about 2.2.18pre15aa1 since they seems completly lfs unrelated). Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
Andrea Arcangeli wrote: > > On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: > > Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches > > applied and some other stuff that we needed, and the machines are stable > > I recommend using 2.2.18pre15aa1 (without using nfsv3 in production for now). Why, specifically, no NFSv3? Do you have a reason or just paranoia? :) I've tested 2.2.18pre15 (w/out 'aa1') against FreeBSD, Solaris, and other Linux clients, and I'm very happy with the improvement over older Linux NFS. All of my testing and real-life usage was done w/ NFSv3 enabled. Jeff - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 04:14:32PM +0200, Andrea Arcangeli wrote: > On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: > > Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli > > patches applied and some other stuff that we needed, and the > > machines are stable > > I recommend using 2.2.18pre15aa1 (without using nfsv3 in production > for now). I'd say that NOT using NFSv3 for production is nothing but folly. The NFSv3 that's in v2.2.18pre15 seems to work VERY fine (we've beaten quite heavily on it without being able to make it barf), and losing all the benefits of NFSv3 in a productionenvironment where NFS is needed isn't an option. I think we should be very lucky that we finally got the NFSv3 client to work, and really encourage people to use it, because Linux has a very bad reputation when it comes to NFS, and it's time to prove to people that we don't suck as much as they believe... :^) /David Weinehall _ _ // David Weinehall <[EMAIL PROTECTED]> /> Northern lights wander \\ // Project MCA Linux hacker// Dance across the winter sky // \> http://www.acc.umu.se/~tao/http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: > Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches > applied and some other stuff that we needed, and the machines are stable I recommend using 2.2.18pre15aa1 (without using nfsv3 in production for now). Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Thu, Oct 12, 2000 at 04:48:55PM -0700, [EMAIL PROTECTED] wrote: > > if you aren't comfortable with dropping a lot of the 2.2.18preX stuff onto > a production box, there is also the 2.2.18pre2aa2 kernel that andrea made 2.2.18pre2aa2 should be kind of rock solid (the only problem it has as every other rawio capable kernel out there is rawio so if you don't use rawio you're just fine for using it in production). The latest aa patchkit (2.2.18pre15aa1) also fixes the couple of severe rawio bugs that can generate memleaks in SMP memory corruption and a deadlock condition that leads to processes unkillable in D state. Obviously nobody before me ever used rawio to read from disk to a completly swapped out area. The swapped out buffer case is the one that two of those bugs trivially (just after one second of runtime the task goes in D state) but if the machine is under a little memory pressure bad things could happen anyways (infact the reason I looked into it the first time was a bugreport about 2.2.18pre2aa2 using a real DBMS on a 256Mbyte RAM machine under some memory pressure, the kernel was returning -EFAULT and writing error messages). The fix also reduces the number of times the kernel walks pagetable for each page that we read/write from/to (it reduces from two walks to 1 walk in the common case where the page is just mapped in memory). I was waiting confirm that there are no problems anymore using DBMS (as I can't definitely reproduce here anymore with my rawio regression test tool) before sending the fix to SCT. But since it's taking some time and I don't want to delay the fixes further I'm extracting the fixes here. 2.4.x needs the most important part of these fixes too. diff -urN rawio-ref/fs/buffer.c rawio/fs/buffer.c --- rawio-ref/fs/buffer.c Fri Oct 13 15:43:59 2000 +++ rawio/fs/buffer.c Fri Oct 13 15:46:06 2000 @@ -1276,6 +1276,7 @@ int i; int err; struct buffer_head *tmp; + unsigned long flags; dprintk ("do_kio start\n"); @@ -1293,9 +1294,10 @@ iosize = 0; } - free_async_buffers(tmp); + put_unused_buffer_head(tmp); iosize += size; } + wake_up(_wait); dprintk ("do_kio end %d %d\n", iosize, err); @@ -1358,6 +1360,7 @@ unsigned long bounce; struct page * map; struct buffer_head *tmp, *bh[KIO_MAX_SECTORS]; + unsigned long flags; /* * First, do some alignment and validity checks @@ -1472,9 +1475,9 @@ error: /* We got an error allocation the bh'es. Just free the current buffer_heads and exit. */ - for (i = bhind; --i >= 0; ) { - free_async_buffers(bh[bhind]); - } + for (i = bhind; --i >= 0; ) + put_unused_buffer_head(bh[i]); + wake_up(_wait); clear_kiobuf_bounce_pages(iobuf); diff -urN rawio-ref/mm/memory.c rawio/mm/memory.c --- rawio-ref/mm/memory.c Fri Oct 13 15:43:59 2000 +++ rawio/mm/memory.c Fri Oct 13 15:46:06 2000 @@ -401,7 +401,7 @@ /* * Do a quick page-table lookup for a single page. */ -static unsigned long get_page(unsigned long address) +static unsigned long get_page(unsigned long address, int write) { pgd_t *pgd; pmd_t *pmd; @@ -411,11 +411,11 @@ if (pmd) { pte_t * pte = pte_offset(pmd, address); if (pte && pte_present(*pte)) { - return pte_page(*pte); + if (!write || pte_write(*pte)) + return pte_page(*pte); } } - printk(KERN_ERR "Missing page in lock_down_page\n"); return 0; } @@ -448,12 +448,15 @@ unsigned long ptr, end; int err; struct mm_struct * mm; - struct vm_area_struct * vma = 0; + struct vm_area_struct * vma = 0, * prev_vma; unsigned long page; struct page * map; int doublepage = 0; int repeat = 0; int i; + int write = (rw == READ); /* if we read from disk +it means we write +to memory */ /* Make sure the iobuf is not already mapped somewhere. */ if (iobuf->nr_pages) @@ -483,16 +486,36 @@ */ while (ptr < end) { if (!vma || ptr >= vma->vm_end) { - vma = find_vma(current->mm, ptr); - if (!vma) + refind: + vma = find_vma_prev(mm, ptr, _vma); + if (!vma) goto out_unlock; + if (vma->vm_start > ptr) { +
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
Andrea Arcangeli wrote: On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches applied and some other stuff that we needed, and the machines are stable I recommend using 2.2.18pre15aa1 (without using nfsv3 in production for now). Why, specifically, no NFSv3? Do you have a reason or just paranoia? :) I've tested 2.2.18pre15 (w/out 'aa1') against FreeBSD, Solaris, and other Linux clients, and I'm very happy with the improvement over older Linux NFS. All of my testing and real-life usage was done w/ NFSv3 enabled. Jeff - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 04:23:51PM +0200, David Weinehall wrote: On Fri, Oct 13, 2000 at 04:14:32PM +0200, Andrea Arcangeli wrote: On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches applied and some other stuff that we needed, and the machines are stable I recommend using 2.2.18pre15aa1 (without using nfsv3 in production for now). I'd say that NOT using NFSv3 for production is nothing but folly. The NFSv3 that's in v2.2.18pre15 seems to work VERY fine (we've beaten I were talking only about 2.2.18pre15aa1. Sorry if I wasn't clear. v2.2.18pre15 has nfsv3 but it's not LFS capable. (however 2.2.18pre15 _may_ have the same problems that I got reported about 2.2.18pre15aa1 since they seems completly lfs unrelated). Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 10:54:50AM -0400, Jeff Garzik wrote: Andrea Arcangeli wrote: On Thu, Oct 12, 2000 at 12:47:39AM +0200, Krzysztof Sierota wrote: Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches applied and some other stuff that we needed, and the machines are stable I recommend using 2.2.18pre15aa1 (without using nfsv3 in production for now). Why, specifically, no NFSv3? Do you have a reason or just paranoia? :) I've a bug report pending for NFSv3 on 2.2.18pre15aa1, it's not paranoia. But I'm not really talking about 2.2.18pre15, it may be a bug in the aa patchkit. OTOH Andi couldn't reproduce it I will try soon too. I've tested 2.2.18pre15 (w/out 'aa1') against FreeBSD, Solaris, and other Linux clients, and I'm very happy with the improvement over older Linux NFS. All of my testing and real-life usage was done w/ NFSv3 enabled. Could you try to use 8192 r/w size? That doesn't happen with the aa patchkit either without changing the packet size. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
Andrea Arcangeli wrote: On Fri, Oct 13, 2000 at 10:54:50AM -0400, Jeff Garzik wrote: I've tested 2.2.18pre15 (w/out 'aa1') against FreeBSD, Solaris, and other Linux clients, and I'm very happy with the improvement over older Linux NFS. All of my testing and real-life usage was done w/ NFSv3 enabled. Could you try to use 8192 r/w size? That doesn't happen with the aa patchkit either without changing the packet size. No problem here, I always use 8192 r/w size :) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Fri, Oct 13, 2000 at 11:20:14AM -0400, Jeff Garzik wrote: No problem here, I always use 8192 r/w size :) You would help me if you could try to do the same with 2.2.18pre15aa1 on both client and server side nfsv3 using also files larger than 2G. Maybe the problem with the bigger r/w size isn't a linux problem but maybe any kind of network problem that triggers increasing the UDP packet size. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
if you aren't comfortable with dropping a lot of the 2.2.18preX stuff onto a production box, there is also the 2.2.18pre2aa2 kernel that andrea made which has the VM stuff. check out andrea/proposed/v2.2/2.2.18pre2 or andrea/kernels/v2.2/2.2.18pre2aa2 unless you've made substantial updates to your patches, andrea? On Mon, 9 Oct 2000, Andrea Arcangeli wrote: > On Mon, Oct 09, 2000 at 04:03:46PM +0100, Alan Cox wrote: > > > The situation persists, we have had two crashes on those servers today. > > > > > > The machines are usually very heavily loaded, 5-50. > > > > > > Please help. > > > Are there any patches to fix this behaviour ? > > > > You might want to try Andrea Arcangeli's further vm patches > > It's here btw: > > >ftp://ftp.us.kernel.org/pub/linux/kernel/people/andrea/patches/v2.2/2.2.18pre9/VM-global-2.2.18pre9-6.bz2 > > It applies cleanly to latest 2.2.18pre. > > Andrea > - > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to [EMAIL PROTECTED] > Please read the FAQ at http://www.tux.org/lkml/ > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Thu, 12 Oct 2000, Krzysztof Sierota wrote: > Our 3 machines went unresponsive, just the way you describe it. The error > was the same. We had this on 2.2.17 and on 2.2.18pre3 , didn't try 2.2.18pre15. > Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches > applied and some other stuff that we needed, and the machines are stable > since then. All our machines run under very _heavy_ disk IO operations. > Loads between 15-30 most of the time. If this is true, I am very much interested in the resultant patch (to vanilla 2.2.17 or 2.2.18pre15) since I have the same load, and the same errors. Could you please send it? On 12 Oct 2000, Per Andreas Buer wrote: > Krzysztof Sierota <[EMAIL PROTECTED]> writes: > > I recently changed the kernel from 2.2.15 to 2.2.17 and added new promise 100 > > card. During 3 days 2 production servers crashed 4 times and had several > > lockups when there was zillion messages like > > VM: do_try_to_free_memmory failed for XXX > Are you using gcc 2.95? We had a lot of trouble (fatal vm-errors) with > gcc 2.95 until we started building kernels with egcs-2.91. Unfortunatelly this is not to blame gcc for: Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/specs gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) -- SaPE Peter, Sasi <[EMAIL PROTECTED]> - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Thu, 12 Oct 2000, Roy C. Bixler wrote: > I just had our mail server running 2.2.18pre15 (compiled with GCC 2.7.2.3) > go unresponcive yesterday. The console was flooded with > 'do_try_to_free_pages failed' messages for various processes and had to be > hard booted to the last stable kernel, which for us is 2.2.16. I noticed > a few minutes before the crash that both 'kupdate' and 'kswapd' were stuck > in the D state and WCHAN for each was 'wait_on_buffer'. I got advice from > Marcelo Tosatti to try out the following Andrea Arcangeli patch: > > ftp://ftp.linux.org/pub/linux/kernel/people/andrea/patches/v2.2/2.2.18pre9/VM-g$ > I feel I can't do that since it's a production machine, but I would be > very interested if your crash sounds anything like ours and if you found a > way to make 2.2.18pre* work on your machine. > Our 3 machines went unresponsive, just the way you describe it. The error was the same. We had this on 2.2.17 and on 2.2.18pre3 , didn't try 2.2.18pre15. Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches applied and some other stuff that we needed, and the machines are stable since then. All our machines run under very _heavy_ disk IO operations. Loads between 15-30 most of the time. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
Krzysztof Sierota <[EMAIL PROTECTED]> writes: > Hi, > I recently changed the kernel from 2.2.15 to 2.2.17 and added new promise 100 > card. During 3 days 2 production servers crashed 4 times and had several > lockups when there was zillion messages like > VM: do_try_to_free_memmory failed for XXX Are you using gcc 2.95? We had a lot of trouble (fatal vm-errors) with gcc 2.95 until we started building kernels with egcs-2.91. -- Per Andreas Buer - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
Krzysztof Sierota [EMAIL PROTECTED] writes: Hi, I recently changed the kernel from 2.2.15 to 2.2.17 and added new promise 100 card. During 3 days 2 production servers crashed 4 times and had several lockups when there was zillion messages like VM: do_try_to_free_memmory failed for XXX Are you using gcc 2.95? We had a lot of trouble (fatal vm-errors) with gcc 2.95 until we started building kernels with egcs-2.91. -- Per Andreas Buer - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Thu, 12 Oct 2000, Roy C. Bixler wrote: I just had our mail server running 2.2.18pre15 (compiled with GCC 2.7.2.3) go unresponcive yesterday. The console was flooded with 'do_try_to_free_pages failed' messages for various processes and had to be hard booted to the last stable kernel, which for us is 2.2.16. I noticed a few minutes before the crash that both 'kupdate' and 'kswapd' were stuck in the D state and WCHAN for each was 'wait_on_buffer'. I got advice from Marcelo Tosatti to try out the following Andrea Arcangeli patch: ftp://ftp.linux.org/pub/linux/kernel/people/andrea/patches/v2.2/2.2.18pre9/VM-g$ I feel I can't do that since it's a production machine, but I would be very interested if your crash sounds anything like ours and if you found a way to make 2.2.18pre* work on your machine. Our 3 machines went unresponsive, just the way you describe it. The error was the same. We had this on 2.2.17 and on 2.2.18pre3 , didn't try 2.2.18pre15. Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches applied and some other stuff that we needed, and the machines are stable since then. All our machines run under very _heavy_ disk IO operations. Loads between 15-30 most of the time. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Thu, 12 Oct 2000, Krzysztof Sierota wrote: Our 3 machines went unresponsive, just the way you describe it. The error was the same. We had this on 2.2.17 and on 2.2.18pre3 , didn't try 2.2.18pre15. Marcelo Tosati assembled a kernel for us that had Andrea Arcangeli patches applied and some other stuff that we needed, and the machines are stable since then. All our machines run under very _heavy_ disk IO operations. Loads between 15-30 most of the time. If this is true, I am very much interested in the resultant patch (to vanilla 2.2.17 or 2.2.18pre15) since I have the same load, and the same errors. Could you please send it? On 12 Oct 2000, Per Andreas Buer wrote: Krzysztof Sierota [EMAIL PROTECTED] writes: I recently changed the kernel from 2.2.15 to 2.2.17 and added new promise 100 card. During 3 days 2 production servers crashed 4 times and had several lockups when there was zillion messages like VM: do_try_to_free_memmory failed for XXX Are you using gcc 2.95? We had a lot of trouble (fatal vm-errors) with gcc 2.95 until we started building kernels with egcs-2.91. Unfortunatelly this is not to blame gcc for: Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/specs gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) -- SaPE Peter, Sasi [EMAIL PROTECTED] - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
if you aren't comfortable with dropping a lot of the 2.2.18preX stuff onto a production box, there is also the 2.2.18pre2aa2 kernel that andrea made which has the VM stuff. check out andrea/proposed/v2.2/2.2.18pre2 or andrea/kernels/v2.2/2.2.18pre2aa2 unless you've made substantial updates to your patches, andrea? On Mon, 9 Oct 2000, Andrea Arcangeli wrote: On Mon, Oct 09, 2000 at 04:03:46PM +0100, Alan Cox wrote: The situation persists, we have had two crashes on those servers today. The machines are usually very heavily loaded, 5-50. Please help. Are there any patches to fix this behaviour ? You might want to try Andrea Arcangeli's further vm patches It's here btw: ftp://ftp.us.kernel.org/pub/linux/kernel/people/andrea/patches/v2.2/2.2.18pre9/VM-global-2.2.18pre9-6.bz2 It applies cleanly to latest 2.2.18pre. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
On Mon, Oct 09, 2000 at 04:03:46PM +0100, Alan Cox wrote: > > The situation persists, we have had two crashes on those servers today. > > > > The machines are usually very heavily loaded, 5-50. > > > > Please help. > > Are there any patches to fix this behaviour ? > > You might want to try Andrea Arcangeli's further vm patches It's here btw: ftp://ftp.us.kernel.org/pub/linux/kernel/people/andrea/patches/v2.2/2.2.18pre9/VM-global-2.2.18pre9-6.bz2 It applies cleanly to latest 2.2.18pre. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: VM: do_try_to_free_memory failed for XXXX, 2.2.17, 2.2.18pre3
> The situation persists, we have had two crashes on those servers today. > > The machines are usually very heavily loaded, 5-50. > > Please help. > Are there any patches to fix this behaviour ? You might want to try Andrea Arcangeli's further vm patches > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/