(I'm resending this, it looks like it was lost somewhere ... and there is some more mem info...) Hello, I have a machine that's dual coppermine/927 (that's what /proc/cpuinfo say:) ), with 1.2G ram, Dell PercRaid, 3com 3c905B NIC. I'm using kernel 2.4.5 with the percraid patches. The problem with the machine is, that it works for some time, and then it goes to hell. I spoke with Rik van Riel last night, and he asked me to get some info from /proc/meminfo around the crash. I've appended it here, and the kernel log messages ( recovered from remote logging station, the machine was unsuble and was rebooted through the power switch :) ) . It doesn't look like the other problems reported with 2.4 VM, so I still haven't tried to use 2.4.6pre3, because it's production machine, and I'd like to stuck with something that's not prerelease :) ( full log is avialable at http://www.ludost.net/eru/ - meminfo.log.bz2, the kernel log from the 'crash' moment, and the dmesg of the machine). (the log was created with while /bin/true; cat /proc/meminfo >> dumpfile;sleep 20; done ). meminfo.log , last lines: total: used: free: shared: buffers: cached: Mem: 1317535744 1308397568 9138176 0 190033920 919121920 Swap: 2657390592 25657344 2631733248 MemTotal: 1286656 kB MemFree: 8924 kB MemShared: 0 kB Buffers: 185580 kB Cached: 897580 kB Active: 973212 kB Inact_dirty: 107736 kB Inact_clean: 2228 kB Inact_target: 744 kB HighTotal: 393208 kB HighFree: 2036 kB LowTotal: 893448 kB LowFree: 6888 kB SwapTotal: 2595108 kB SwapFree: 2570052 kB total: used: free: shared: buffers: cached: Mem: 1317535744 1312010240 5525504 0 184389632 927334400 Swap: 2657390592 25657344 2631733248 MemTotal: 1286656 kB MemFree: 5396 kB MemShared: 0 kB Buffers: 180068 kB Cached: 905600 kB Active: 980272 kB Inact_dirty: 103148 kB Inact_clean: 2248 kB Inact_target: 748 kB HighTotal: 393208 kB HighFree: 2036 kB LowTotal: 893448 kB LowFree: 3360 kB SwapTotal: 2595108 kB SwapFree: 2570052 kB total: used: free: shared: buffers: cached: Mem: 1317535744 1312051200 5484544 0 175587328 936456192 Swap: 2657390592 25657344 2631733248 MemTotal: 1286656 kB MemFree: 5356 kB MemShared: 0 kB Buffers: 171472 kB Cached: 914508 kB Active: 985180 kB Inact_dirty: 98440 kB Inact_clean: 2368 kB Inact_target: 712 kB HighTotal: 393208 kB HighFree: 2036 kB LowTotal: 893448 kB LowFree: 3320 kB SwapTotal: 2595108 kB SwapFree: 2570052 kB total: used: free: shared: buffers: cached: Mem: 1317535744 1312182272 5353472 0 167284736 943894528 Swap: 2657390592 25657344 2631733248 MemTotal: 1286656 kB MemFree: 5228 kB MemShared: 0 kB Buffers: 163364 kB Cached: 921772 kB Active: 988760 kB Inact_dirty: 93888 kB Inact_clean: 2488 kB Inact_target: 772 kB HighTotal: 393208 kB HighFree: 2036 kB LowTotal: 893448 kB LowFree: 3192 kB SwapTotal: 2595108 kB SwapFree: 2570052 kB kern.log , last lines : Jun 23 08:17:41 eru kernel: possible SYN flooding on port 80. Sending cookies. Jun 23 08:17:41 eru kernel: __alloc_pages: 1-order allocation failed. Jun 23 08:18:41 eru last message repeated 3201 times Jun 23 08:18:41 eru kernel: possible SYN flooding on port 80. Sending cookies. Jun 23 08:18:41 eru kernel: __alloc_pages: 1-order allocation failed. Jun 23 08:19:41 eru last message repeated 3191 times Jun 23 08:19:41 eru kernel: possible SYN flooding on port 80. Sending cookies. Jun 23 08:19:41 eru kernel: __alloc_pages: 1-order allocation failed. Jun 23 08:20:41 eru kernel: possible SYN flooding on port 80. Sending cookies. Jun 23 08:20:41 eru last message repeated 3354 times Jun 23 08:20:42 eru kernel: __alloc_pages: 1-order allocation failed. And, meminfo from the last crash ( happened 30 mins ago ) : total: used: free: shared: buffers: cached: Mem: 1317535744 1310912512 6623232 0 142184448 897589248 Swap: 2657390592 2813952 2654576640 MemTotal: 1286656 kB MemFree: 6468 kB MemShared: 0 kB Buffers: 138852 kB Cached: 876552 kB Active: 765348 kB Inact_dirty: 248260 kB Inact_clean: 1796 kB Inact_target: 300 kB HighTotal: 393208 kB HighFree: 2036 kB LowTotal: 893448 kB LowFree: 4432 kB SwapTotal: 2595108 kB SwapFree: 2592360 kB - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/