Re: any advice to find root cause of Falling back to HPET ?
On Sun, 2010-05-23 at 21:11 -0300, Andrew Reid wrote: A few random thoughts: Any chance the RAID is still rebuilding. Is slow unit truly slower, that is does it take longer as YOU measure it to do a significant task. I ask because a bad clock will would make it report longer times due to time measurement errors. A comparison of the actual number of interrupts (cat /proc/interrupts) on both machines after they have been up for similar periods, doing similar things may give a hint as to who has 'stolen' your CPU. Yes, tasks do take longer in real time. (a lot longer) The RAID was not rebuilding. The 2nd chassis started behaving strangely, and upon reboot would not finish booting. The guys at the hosting company said the server was still booting up after 15 minutes and going VERY VERY slowly. They finally gave up and decided to swap chassis again (on 3rd chassis now). This time around, on 3rd chassis, the lost ticks are no longer, the server is fast/normal again, and all is well. I just can't believe we ran into 2 servers with the same issue back to back??? (the guys at the hosting company are usually great, but it almost makes me doubt that they did the 1st chassis swap??) Anyway, all is well again bizarre weekend indeed. -Bond ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
Re: any advice to find root cause of Falling back to HPET ?
On 2010-05-24 01:01, Bond Masuda wrote: This time around, on 3rd chassis, the lost ticks are no longer, the server is fast/normal again, and all is well. I just can't believe we ran into 2 servers with the same issue back to back??? (the guys at the hosting company are usually great, but it almost makes me doubt that they did the 1st chassis swap??) For future reference, you might want to save a dump of dmidecode output before requesting chassis swaps so you can check if the service tag actually changed. ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
RE: any advice to find root cause of Falling back to HPET ?
Thanks. Good point. As an aside, this weekend has been a bizarre weekend. After finally resolving this lost ticks issue, another server kernel panic and crashed mysteriously, and again kernel panic upon boot up.. some messages about bad memory in DIMM8 and DIMM7. then my friend's Drobo went all red and failed this afternoon. We must be getting showered by intense cosmic rays this weekend thanks for the responses from everyone. -Bond -Original Message- From: linux-poweredge-boun...@dell.com [mailto:linux-poweredge- boun...@dell.com] On Behalf Of Jefferson Ogata Sent: Sunday, May 23, 2010 9:01 PM To: linux-powere...@lists.us.dell.com Subject: Re: any advice to find root cause of Falling back to HPET ? On 2010-05-24 01:01, Bond Masuda wrote: This time around, on 3rd chassis, the lost ticks are no longer, the server is fast/normal again, and all is well. I just can't believe we ran into 2 servers with the same issue back to back??? (the guys at the hosting company are usually great, but it almost makes me doubt that they did the 1st chassis swap??) For future reference, you might want to save a dump of dmidecode output before requesting chassis swaps so you can check if the service tag actually changed. ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
Re: any advice to find root cause of Falling back to HPET ?
On 2010-05-24 04:45, Bond Masuda wrote: As an aside, this weekend has been a bizarre weekend. After finally resolving this lost ticks issue, another server kernel panic and crashed mysteriously, and again kernel panic upon boot up.. some messages about bad memory in DIMM8 and DIMM7. then my friend's Drobo went all red and failed this afternoon. We must be getting showered by intense cosmic rays this weekend Maybe the lost ticks was happening because of the Lost series finale. rim-shot / ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
any advice to find root cause of Falling back to HPET ?
Hello, I'd appreciate any help/advice anyone can provide regarding our issue. I've run out of ideas on this one... We have two identical PowerEdge 2950, one is called s7 and the other is s8. Both are web servers running Apache and PHP. We first noticed the problem because our benchmarking showed drastically different results between the two servers. With s7, we were able to get 180 requests/sec while on s8 we only get 35 request/sec (and now only 15 requests/sec - more on that below). After this, we became aware that almost all tasks on s8 were slower than s7, whether it is CPU bound or I/O bound, everything we tried was slower on s8 than on s7 (untar'ing archives, running md5 hashes, etc). I started digging around. Both servers are identical in terms of software and configuration (other than things like hostname and IP addresses). Both servers are RHEL4U8, kernel-2.6.9-89.0.25.ELsmp, x86_64, exact same packages and exact same versions. I even ran 'rpm --verify' on all packages and didn't find anything unusual on both s7 and s8. The ONLY error message I'm seeing that is unique to s8 are the following messages in dmesg: Losing some ticks... checking if CPU frequency changed. warning: many lost ticks. Your time source seems to be instable or some driver is hogging interupts rip __do_softirq+0x4d/0xd0 Falling back to HPET Some google searching found: https://bugzilla.redhat.com/show_bug.cgi?id=429010 which refers to: https://bugzilla.redhat.com/show_bug.cgi?id=248488 But that seems to refer to problems with virtualization. This is on real hardware. What we don't understand is that s7 does *not* exhibit any slowness nor the messags above, only s8. Again, both are identical. So, thinking this might be a hardware issue, we asked our hosting company to pull the drives out of s8 and replace the entire chassis. After replacing the entire chassis of s8, we are still getting the above messages in dmesg. Not only that, things have gotten worse... our benchmarking (using 'ab') now shows the server can only do 15 requests/sec (all these test were run locally on loopback to avoid any network related issue). Since the chassis was swapped, we feel that it probably isn't a hardware issue. But we have s7 which is configured identically to s8 that doesn't have this issue, so it is hard to say that it is a software issue. Any advice? What can I do to find the root cause? TIA, -Bond ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
RE: any advice to find root cause of Falling back to HPET ?
thanks for the reply. I'm pretty sure it is not #2, we have verified the software on both s7 and s8 are identical. it is true, the only hardware component that we kept using are the hard drives when the server chassis was swapped out. I just don't understand how a faulty drive can cause many lost ticks and those other messages. Nonetheless, I'm not going to exclude the possibility. I wish RHEL4 had smartctl that worked with megasas; i'll have to compile the latest smartctl to see if SMART data will tell me anything about the drives. One thing to note is that during the build of these servers 2 weeks ago, one of the drives on s8 did fail and had to be replaced. -Bond -Original Message- From: guy [mailto:guy.choo.ke...@gmail.com] Sent: Saturday, May 22, 2010 12:42 PM To: Bond Masuda Cc: linux-powere...@lists.us.dell.com Subject: Re: any advice to find root cause of Falling back to HPET ? you can try putting s8's drives inside s7 and see what you get. if you get errors, this can be one of: 1. the hard disk themselves are faulty. or 2. there is some different driver on s8 that is not found on s7 (or the other way around) and it was installed not via the RPM system (so in RPM you won't see a difference) - which is causing the problems. --guy Bond Masuda wrote: Hello, I'd appreciate any help/advice anyone can provide regarding our issue. I've run out of ideas on this one... We have two identical PowerEdge 2950, one is called s7 and the other is s8. Both are web servers running Apache and PHP. We first noticed the problem because our benchmarking showed drastically different results between the two servers. With s7, we were able to get 180 requests/sec while on s8 we only get 35 request/sec (and now only 15 requests/sec - more on that below). After this, we became aware that almost all tasks on s8 were slower than s7, whether it is CPU bound or I/O bound, everything we tried was slower on s8 than on s7 (untar'ing archives, running md5 hashes, etc). I started digging around. Both servers are identical in terms of software and configuration (other than things like hostname and IP addresses). Both servers are RHEL4U8, kernel-2.6.9-89.0.25.ELsmp, x86_64, exact same packages and exact same versions. I even ran 'rpm --verify' on all packages and didn't find anything unusual on both s7 and s8. The ONLY error message I'm seeing that is unique to s8 are the following messages in dmesg: Losing some ticks... checking if CPU frequency changed. warning: many lost ticks. Your time source seems to be instable or some driver is hogging interupts rip __do_softirq+0x4d/0xd0 Falling back to HPET Some google searching found: https://bugzilla.redhat.com/show_bug.cgi?id=429010 which refers to: https://bugzilla.redhat.com/show_bug.cgi?id=248488 But that seems to refer to problems with virtualization. This is on real hardware. What we don't understand is that s7 does *not* exhibit any slowness nor the messags above, only s8. Again, both are identical. So, thinking this might be a hardware issue, we asked our hosting company to pull the drives out of s8 and replace the entire chassis. After replacing the entire chassis of s8, we are still getting the above messages in dmesg. Not only that, things have gotten worse... our benchmarking (using 'ab') now shows the server can only do 15 requests/sec (all these test were run locally on loopback to avoid any network related issue). Since the chassis was swapped, we feel that it probably isn't a hardware issue. But we have s7 which is configured identically to s8 that doesn't have this issue, so it is hard to say that it is a software issue. Any advice? What can I do to find the root cause? TIA, -Bond ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
Re: any advice to find root cause of Falling back to HPET ?
On 2010-05-22 19:52, Bond Masuda wrote: I wish RHEL4 had smartctl that worked with megasas; i'll have to compile the latest smartctl to see if SMART data will tell me anything about the drives. One thing to note is that during the build of these servers 2 weeks ago, one of the drives on s8 did fail and had to be replaced. Try megactl/megasasctl. ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq