Re: [CentOS] Network bond - one port goes down from time to time
Am 29.03.16 um 13:57 schrieb Marcelo Ricardo Leitner: > Em 29-03-2016 03:46, Götz Reinicke - IT Koordinator escreveu: >> Am 28.03.16 um 16:23 schrieb Marcelo Ricardo Leitner: >>> Em 28-03-2016 06:27, Götz Reinicke escreveu: Hi, may be someone has an idea: We have three supermicron servers with two 10Gb Ports each, connected to a cisco switch stack 1Gb ports. All are on auto speed. I configured a LACP bond on both sides on all servers, first with citrix xen server. On one server eth0 goes down from time to time … maybe within minutes, someday it is up for some hours. Two server are fine; the bond is up for 24 days(!) now without any problem. Recently I installed centos 7.2 on that server in question and - bam - eth0 is going down from time to time … I checked patch cables, tried an other switch port channel, reconfigured the ports, reinstalled the os. Same behavior. And: We got a replacement server. Same behavior …. :) Currently the cisco tech guys don’t see a problem on the switch (which is up for 3 Years now with 10+ servers connected … no problem so far), from the citrix side I don’t get much more hints. In the logs i just have a Nic Link is Down … Nic Link is Up. It is always eth0. Question: Any idea ? One suggestion was Disable all power saving features in the server bios. Did not do that yet. Is there any chance to set some sort of higher debug level for that nic/kernel/whatever to get some server os side feedback why the port goes down? Regards and thanks for any hint! . Götz >>> >>> If you are seeing NIC Link is Down as in: >>> [710442.668059] e1000e: enp0s25 NIC Link is Down >>> then the NIC lost its link and bond is just protecting you as you >>> probably didn't have any downtime due to that. IOW bonding is not the >>> issue. >>> >>> Which NIC do you have on those servers? >> >> >> The mainbord is a supermicro X10DRI-T with Intel X540 Dual port >> 10GBase-T. > > Okay, it's probably using ixgbe driver then. > You may consider testing a newer kernel and see how that goes out, > before doing too much debugging. > You can install v4.5 using one of ELrepo's kernels at > http://elrepo.org/linux/kernel/el7/x86_64/RPMS/ > http://elrepo.org/tiki/tiki-index.php > There are some changes between 7.2 and that kernel that it's good to be > tested. > > Or... enable ixgbe debug, module param debug=16, and send the dmesg log, > specially the lines around the event. Hm,, could you give me a hint, how to enable that (at runtime) for centos 7.2? I cant figure that out. Would be nice. cheers . Götz ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS6: Building a modified kernel rpm
Dear Akemi, I'm really ashamed, in "https://wiki.centos.org/HowTos/I_need_the_Kernel_Source"; I've missed the last 5 lines following the CentOS5 section. All is now clear for me on the process to follow to build a new kernel from source rpm and the CONFIG_NUMA_EMU option is active. I've set it with make menuconfig to avoid some additional mistakes in the config file. However, with this CONFIG_NUMA_EMU option , the compiler returns an error and compilation fails. arch/x86/mm/srat_64.c: In function 'acpi_fake_nodes': arch/x86/mm/srat_64.c:469: error: implicit declaration of function '__acpi_map_pxm_to_node' make[2]: *** [arch/x86/mm/srat_64.o] Error 1 make[1]: *** [arch/x86/mm] Error 2 make: *** [arch/x86] Error 2 make: *** Waiting for unfinished jobs error: Bad exit status from /var/tmp/rpm-tmp.VkJ805 (%build) Bad exit status from /var/tmp/rpm-tmp.VkJ805 (%build) Without CONFIG_NUMA_EMU option, compilation is successfull. It is the CentOSplus kernel from kernel-2.6.32-573.el6.centos.plus.src.rpm as yous suggested. Thank for your patience Patrick Akemi Yagi wrote: On Tue, Mar 29, 2016 at 2:27 AM, Patrick Begou < patrick.be...@legi.grenoble-inp.fr> wrote: Akemi Yagi wrote: I recommend you follows the instructions in this CentOS wiki article: https://wiki.centos.org/HowTos/Custom_Kernel If you encounter any issue, please let us know here. Akemi Thanks for answering Akemi . Indeed I tried using this howto before posting but something is wrong in this tutorial. I installed the prerequisites and downloaded the kernel sources as detailed in section 2 of https://wiki.centos.org/HowTos/I_need_the_Kernel_Source. But on https://wiki.centos.org/HowTos/Custom_Kernel, section 2, all goes wrong! My ~/rpmbuild/BUILD directory is empty! No configs directory too. No ~/rpmbuild/BUILD/kernel-*/linux-*/ available, all configs files are in ~/rpmbuild/SOURCES The referenced wiki article should work for building CentOS-6 kernels (not quite ready for CentOS-7). You really have to dot the i's and cross the t's. If you miss a step or command in there, chances are it will fail. Since this I try to modify ~/rpmbuild/SOURCES/config-x86-generic because it was the only file containing the CONFIG_ACPI_NUMA flag. No success, the build does not take account of my changes as the ~/rpmbuild/BUILD/kernel-2.6.32-573.22.1.el6/linux-2.6.32-573.22.1.el6.numa.x86_64/.config created file with: rpmbuild -bb --without xen --without debug --without debuginfo --target=x86_64 ./SPECS/kernel.spec says "# CONFIG_NUMA_EMU is not set" So I change this file with an editor and rebuild the rpm package of the kernel with the same command. The .config file is overwriten. As you found out, editing the existing config-x86-generic file would not work because the final config file is created on the fly. This is why the wiki instructions tell you to create your customized config file and show you how to get that config file to be used during the build process. I have another suggestion for you. Get the srpm for the CentOSPlus kernel instead of the distro kernel. The plus kernel is a customized kernel. It has kernel-xx-config files that you can directly edit. They will not be overwritten. Perhaps this method is easier. But if you wish to go for the original method, just be extremely precise not to skip any step. Hope this helps, Akemi ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos -- === | Equipe M.O.S.T. | | | Patrick BEGOU | mailto:patrick.be...@grenoble-inp.fr | | LEGI| | | BP 53 X | Tel 04 76 82 51 35 | | 38041 GRENOBLE CEDEX| Fax 04 76 82 52 71 | === ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] CentOS6: Building a modified kernel rpm
Patrick Begou wrote: Dear Akemi, I'm really ashamed, in "https://wiki.centos.org/HowTos/I_need_the_Kernel_Source"; I've missed the last 5 lines following the CentOS5 section. All is now clear for me on the process to follow to build a new kernel from source rpm and the CONFIG_NUMA_EMU option is active. I've set it with make menuconfig to avoid some additional mistakes in the config file. However, with this CONFIG_NUMA_EMU option , the compiler returns an error and compilation fails. arch/x86/mm/srat_64.c: In function 'acpi_fake_nodes': arch/x86/mm/srat_64.c:469: error: implicit declaration of function '__acpi_map_pxm_to_node' make[2]: *** [arch/x86/mm/srat_64.o] Error 1 make[1]: *** [arch/x86/mm] Error 2 make: *** [arch/x86] Error 2 make: *** Waiting for unfinished jobs error: Bad exit status from /var/tmp/rpm-tmp.VkJ805 (%build) Bad exit status from /var/tmp/rpm-tmp.VkJ805 (%build) Without CONFIG_NUMA_EMU option, compilation is successfull. It is the CentOSplus kernel from kernel-2.6.32-573.el6.centos.plus.src.rpm as yous suggested. Redhat make very big changes to the base kernel source to create their kernel - meaning that code they don't use (i.e. code included with CONFIG options they don't support) may no longer build - as is probably the case here You may want to compare the 2.6.32-573 source with the generic 2.6.32 kernel source where your compile fails above to see if you can find a workaround A useful starting point could be: http://lxr.free-electrons.com/ident?v=2.6.32;i=__acpi_map_pxm_to_node However, there is no guarantee that even if you get it to compile that it will work as expected ... James Pearson ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Network bond - one port goes down from time to time
Em 30-03-2016 06:46, Götz Reinicke - IT Koordinator escreveu: Am 29.03.16 um 13:57 schrieb Marcelo Ricardo Leitner: Em 29-03-2016 03:46, Götz Reinicke - IT Koordinator escreveu: Am 28.03.16 um 16:23 schrieb Marcelo Ricardo Leitner: Em 28-03-2016 06:27, Götz Reinicke escreveu: Hi, may be someone has an idea: We have three supermicron servers with two 10Gb Ports each, connected to a cisco switch stack 1Gb ports. All are on auto speed. I configured a LACP bond on both sides on all servers, first with citrix xen server. On one server eth0 goes down from time to time … maybe within minutes, someday it is up for some hours. Two server are fine; the bond is up for 24 days(!) now without any problem. Recently I installed centos 7.2 on that server in question and - bam - eth0 is going down from time to time … I checked patch cables, tried an other switch port channel, reconfigured the ports, reinstalled the os. Same behavior. And: We got a replacement server. Same behavior …. :) Currently the cisco tech guys don’t see a problem on the switch (which is up for 3 Years now with 10+ servers connected … no problem so far), from the citrix side I don’t get much more hints. In the logs i just have a Nic Link is Down … Nic Link is Up. It is always eth0. Question: Any idea ? One suggestion was Disable all power saving features in the server bios. Did not do that yet. Is there any chance to set some sort of higher debug level for that nic/kernel/whatever to get some server os side feedback why the port goes down? Regards and thanks for any hint! . Götz If you are seeing NIC Link is Down as in: [710442.668059] e1000e: enp0s25 NIC Link is Down then the NIC lost its link and bond is just protecting you as you probably didn't have any downtime due to that. IOW bonding is not the issue. Which NIC do you have on those servers? The mainbord is a supermicro X10DRI-T with Intel X540 Dual port 10GBase-T. Okay, it's probably using ixgbe driver then. You may consider testing a newer kernel and see how that goes out, before doing too much debugging. You can install v4.5 using one of ELrepo's kernels at http://elrepo.org/linux/kernel/el7/x86_64/RPMS/ http://elrepo.org/tiki/tiki-index.php There are some changes between 7.2 and that kernel that it's good to be tested. Or... enable ixgbe debug, module param debug=16, and send the dmesg log, specially the lines around the event. Hm,, could you give me a hint, how to enable that (at runtime) for centos 7.2? I cant figure that out. Would be nice. cheers . Götz Ah during runtime you can just use ethtool: # ethtool -s eth0 msglvl 0x when done, revert with: # ethtool -s eth0 msglvl 0x7 Marcelo ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
[CentOS] Is there a way to detect/validate DHCP static IP assignment?
Hi, We have tens of networks(VLANs) in data center with a central Linux DHCP server. each network has their router to do the DHCP relay. So, the DHCP server's configuration files has tens 'subnet' statements. Because PXE booting is standard in whole data center, there are also thousands of static MAC-IP mapping 'host' statements in dhcp configuration. The big challenge with a central dhcp server is how to detect typo in the thousands of MAC-IP 'host' statements? -- a single char/digit typo here will fail a PXE booting or download wrong post-installation snippets. Is there a tool to validate all 'host' statements from another Linux box? I tried nagios check-dhcp plugin, with a series of real MAC addresses(for hosts in other different networks). Surprisingly, the IP address came back were not the static IP addresses in 'host' statement, but dynamic addresses in the pool defined for this particular network (where I ran check-dhcp from). check_dhcp was run with the following arguments: /usr/lib64/nagios/plugins/check_dhcp --verbose --server= --interface=eth0 --mac= --unicast remove --unicast doesn't help but just see more DHCP replys. Interestingly, with a same MAC address, and the above same command, from two Centos boxes on different network there will be different dynamic IPs! instead the static IP defined with 'host' statement. So, how can we validate static IP assignment? Thanks. Best,David, ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Centos in the Browser string ?
> -Original Message- > From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On > Behalf Of g > Sent: den 24 mars 2016 16:56 > To: centos@centos.org > Subject: Re: [CentOS] Centos in the Browser string ? > > if you have a file in profile directory, add this to it. if not, > create file and paste this in it. > > //set user agent to blank > user_pref("general.useragent.override", " "); > > what makes you get a unique rating is that you report no agent. only > info any site will know about you is your ip address. > > if you want to hide that, use a proxy server. ((GBWG)) Please excuse my ignorance, but what file in what profile directory is referred to here? -- //Sorin ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Centos in the Browser string ?
> -Original Message- > From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On > Behalf Of g > Sent: den 24 mars 2016 17:22 > To: centos@centos.org > Subject: Re: [CentOS] Centos in the Browser string ? > > > now i goofed. :-\ > > > if you have file "user.js" in profile directory, Ah, thank you! -- //Sorin ___ CentOS mailing list CentOS@centos.org https://lists.centos.org/mailman/listinfo/centos