Re: [CentOS] CentOS 6 can't custom partition layout in text modeinstall
On Tue, Jul 12, 2011 at 01:24:25PM +0100, Keith Roberts wrote: On Tue, 12 Jul 2011, David wrote: To: CentOS mailing list centos@centos.org From: David da...@pnyet.web.id Subject: [CentOS] CentOS 6 can't custom partition layout in text mode install Dear All, Today I'm trying to install CentOS 6 boot using text mode, but I can't create custom partition layout because I can't find the custom partition layout menu. Where is manually partition setup in text mode install? Please advice... Or, you could just use Kickstart... Example: zerombr part /boot --fstype=ext4 --size=128 --ondisk=sda part pv.01 --size=100 --grow --ondisk=sda volgroup vg01 pv.01 logvol / --fstype ext4 --name=rootlv --vgname=vg01 --size=2048 logvol /tmp --fstype ext4 --name=tmplv --vgname=vg01 --size=384 logvol /usr --fstype ext4 --name=usrlv --vgname=vg01 --size=8192 logvol /var --fstype ext4 --name=varlv --vgname=vg01 --size=2048 logvol /opt --fstype ext4 --name=optlv --vgname=vg01 --size=1024 logvol /home --fstype ext4 --name=homelv --vgname=vg01 --size=2048 logvol swap --fstype swap --name=swaplv --vgname=vg01 --recommended Good luck, Josh ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Centos 6 Server has no GUI
on Tue, 12 Jul 2011, Michael Schumacher wrote: To: CentOS mailing list centos@centos.org From: Michael Schumacher michael.schumac...@pamas.de Subject: Re: [CentOS] Centos 6 Server has no GUI Dear Giles, On Monday, July 11, 2011 you wrote: So first daft question with Centos 6 (someone had to be first!) I've setup Centos 6 as a Server but as with Centos 5 it used to boot into the GUI but v6 doesn't do this, startx etc doesn't seem to work to launch the GUI perhaps:: yum groupinstall GNOME Desktop Environment There is no GNOME Desktop Environment group. Check with yum grouplist and you will see. And even installing KDE Desktop won't help, because it is not starting automatically. And then, we don't have runlevels any more, this means a little bit more looking into details is required. How can Linux _not_ have run levels. I thought that was a central part of the design of Linux? Um, who says 6 doesn't have runlevels? I haven't personally looked at C6, but upstream EL6 -does- have runlevels just as previous versions: # 0 - halt (Do NOT set initdefault to this) # 1 - Single user mode # 2 - Multiuser, without NFS (The same as 3, if you do not have networking) # 3 - Full multiuser mode # 4 - unused # 5 - X11 # 6 - reboot (Do NOT set initdefault to this) Josh ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Setting hostname and network settings in 6.0 text mode installer
On Tue, 2011-07-12 at 15:14 -0600, Joe Mata wrote: I'm doing my first 6.0 install the same way as I've always done my server installs -- in text mode. But it looks like there's no option to do the following in the 6.0 text mode installer? * Set hostname * Set network settings * Choose packages I really hope I'm missing something because having to use a GUI to install a server is incredibly painful. Try using Kickstart, especially if you are deploying multiple servers. Believe me, it makes life MUCH easier. Josh ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] SNMP monitoring options
I'd give ZenOSS a try if I were you. -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Les Mikesell Sent: Wednesday, April 13, 2011 1:50 PM To: centos@centos.org Subject: Re: [CentOS] SNMP monitoring options On 4/13/2011 12:39 PM, aurfal...@gmail.com wrote: We looked at a number of monitoring systems before closing on (community version) Opsview. The main other contender we looked at was zenoss - I have to admit I was biased due to previous use of and liking of nagios. So, if I already have a working nagios that I like, are there any advantages to Opsview community? I only have about 100 services I monitor, and I doubt it'd grow a huge amount in the near future. Well, Nagios is more for outage/alerts were I was looking more for trending. Thats why I was originally using Cacti w/Nagios plugin. If you find anything that is really good at this, please post the info. I'm trying to do it with OpenNMS, but the problem is that the trends I want to track are always across load balanced/fail over sets of things and the aggregation needs to be done at each sample interval to get it right (i.e. if you fail over between two routers you can't add the peak usage of both over a longer interval and call it a trend). I've ended up exporting most of the data out to other tools for trend analysis. -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Auto-updates -- Bad Idea?
If you don't want your system to break unexpectedly, do not enable automatic updates; especially if you are running any packages with non-standard configurations. In some cases, with bare packages and stock configurations, automatic updates may prove to be a viable and SEMI-safe solution. But, if you wish to maintain the integrity of a production system, I would suggest that you do not use them. Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of email builder Sent: Wednesday, April 06, 2011 2:36 PM To: centos@centos.org Subject: [CentOS] Auto-updates -- Bad Idea? Hello, As I've learned recently, I do not have any auto updates configured on my system. I see some posts on the web encouraging the use of yum-cron, but I'd like to know what people feel about the use of automatic updates. That is, for a server (non-desktop) system, automatic updates could break things or have other unforeseen consequences, and that could happen at the worst of times, since the process runs regularly. On the other hand, for small businesses without highly trained sysadmins or ones with enough time to baby their servers, missing critical updates to, say openssl or some other mission-critical package could spell disaster. Is the only reasonable solution to schedule a human cron once a week to look at needed updates? Ouch. Thanks in advance for your considered opinions. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Affordable KVM over IP switch
You can pick up a Dell/Avocent 2161DS on eBay for $400-500 USD. -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Michael B Allen Sent: Monday, March 21, 2011 11:16 PM To: CentOS Subject: [CentOS] Affordable KVM over IP switch Hello, Are there any KVM over IP switches that are not thousands of dollars? Ideally a 3-4 port switch for a few hundred seems reasonable to me. Mike -- Michael B Allen Java Active Directory Integration http://www.ioplex.com/ ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Recommendation for a Good Vulnerability Scanning Service?
We use Qualys for PCI vulnerability scanning. Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Michael B Allen Sent: Friday, February 18, 2011 1:20 PM To: centos@centos.org Subject: [CentOS] Recommendation for a Good Vulnerability Scanning Service? Hi, Can someone recommend a good vulnerability scanning service? I just need the minimum for PCI compliance (it's a sort of credit card processing certification). I got a free scan from https://www.hackerguardian.com/ and their scan reported a number of Fail results. I haven't checked them all yet but most seem to be things for which fixes were backported looong ago by The Upstream Vendor. I haven't spoken with the hackerguardian people yet but it would be nice if I could just say I'm using CentOS 5.5 and have them factor that into their report so that I can focus on any real issues. Are there vulnerability scanning services that are more or less sophisticated about this? Thanks, Mike ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] sudo doing DNS lookup
Maybe I am missing something here.. but what does 'sudo' have to do with DNS resolution? From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Steve Clark Sent: Friday, December 10, 2010 7:44 AM To: CentOS mailing list Subject: [CentOS] sudo doing DNS lookup Hi, I have a confusing problem. I have two centos 5,5 boxes. Both have sudo.i3861.7.2p1-9.el5_5 installed I am using the same sudoers file, but the one on box A keeps trying to do DNS lookups while the one on box B does not. How do I disable this DNS lookup? Thanks for any info. -- Stephen Clark NetWolves Sr. Software Engineer III Phone: 813-579-3200 Fax: 813-882-0209 Email: steve.cl...@netwolves.com http://www.netwolves.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] sendmail substitute?
Actually, as of RHEL6, the default MTA is now Postfix. Sendmail does indeed have a rather lengthy history of vulnerabilities. With that being said, in my opinion, Postfix is also a much more flexible MTA. Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Morten P.D. Stevens Sent: Thursday, October 14, 2010 1:55 PM To: CentOS mailing list Subject: Re: [CentOS] sendmail substitute? On Thu, Oct 14, 2010 at 9:11 AM, Alexander Dalloz ad+li...@uni-x.org wrote: What is the rationale behind deactivating Sendmail. Just curious. Or is it the typical rant Sendmail is insecure, see its history? I don't understand why many people calling sendmail insecure. Sendmail is the default MTA in RHEL, Solaris, AIX, FreeBSD, OpenBSD ... Why should they use an insecure MTA? Sendmail is a very robust and reliable MTA. Best regards, Morten ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] question on modprobe.conf
The original poster's question was referring to RHEL 6b2, not RHEL/Cent 5.5 which still uses /etc/modprobe.conf that you are used to. Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of ken Sent: Wednesday, July 07, 2010 6:48 AM To: CentOS mailing list Subject: Re: [CentOS] question on modprobe.conf On 07/06/2010 02:54 PM Jerry Geis wrote: I download the RHEL 6b2 and noticed the modprobe.conf was no longer present. I used modprobe.conf to place options for ethernet drivers like which order to install the drivers. e1000e before forcedeth - things like that. Anyway - seems like that file is no longer present. what is the future way to handle driver options? Thanks Jerry On this system (with the latest 5.5 updates) there is an /etc/modprobe.conf file. It must have been created by the system... because I didn't create it. Nor have I made any changes to it... ever. Moreover, the Access, Modify, and Change timestamps all correspond to the last time the system was booted. According to man modprobe.conf: There is a generate_modprobe.conf program which should do a reasonable job of generating modprobe.conf from your current (2.4 or 2.2) modules setup. This program, however, isn't present on my system. Manpages should have dates in the content revealing their last update. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] question on modprobe.conf
I'm guessing alphabetically? From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of cornel panceac Sent: Tuesday, July 06, 2010 2:26 PM To: CentOS mailing list Subject: Re: [CentOS] question on modprobe.conf Similar content is now broken up into separate .conf files in directory /etc/modprobe.d so that individual packages can now have sole ownership of a file rather than trying to pack all their parameters into a single, hard to maintain file. in what order are the files processed? -- Among the maxims on Lord Naoshige's wall, there was this one: Matters of great concern should be treated lightly. Master Ittei commented, Matters of small concern should be treated seriously. (Ghost Dog : The Way of The Samurai) ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Yum archive for PHP greater than 5.2.1 for CentOS 5.4?
Check the centosextra's repo. Josh From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of kOoLiNuS Sent: Monday, June 28, 2010 10:54 AM To: CentOS mailing list Subject: Re: [CentOS] Yum archive for PHP greater than 5.2.1 for CentOS 5.4? Joseph L. Casale wrote: Whether I like it or not I need to get a version of PHP that is greater than 5.2.1 http://blog.famillecollet.com/pages/Config-en +1 with EPEL this is what I use on my systems too (but disable the repo after installation since they update fast) -- Nicola Losito.:kOoLiNuS:. http://claimid.com/koolinus ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Yum archive for PHP greater than 5.2.1 for CentOS 5.4?
I'm sorry, I meant the 'c5-testing' repo. From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Baird, Josh Sent: Monday, June 28, 2010 10:57 AM To: CentOS mailing list Subject: Re: [CentOS] Yum archive for PHP greater than 5.2.1 for CentOS 5.4? Check the centosextra's repo. Josh From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of kOoLiNuS Sent: Monday, June 28, 2010 10:54 AM To: CentOS mailing list Subject: Re: [CentOS] Yum archive for PHP greater than 5.2.1 for CentOS 5.4? Joseph L. Casale wrote: Whether I like it or not I need to get a version of PHP that is greater than 5.2.1 http://blog.famillecollet.com/pages/Config-en +1 with EPEL this is what I use on my systems too (but disable the repo after installation since they update fast) -- Nicola Losito.:kOoLiNuS:. http://claimid.com/koolinus ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] 2 DVDs for CentOS 5.5
DVD2 only contains OpenOffice packages. Unless you need to install these from the DVD, you should be fine without DVD 2. Josh From: centos-boun...@centos.org on behalf of David McGuffey Sent: Tue 6/22/2010 9:46 PM To: CentOS mailing list Subject: [CentOS] 2 DVDs for CentOS 5.5 Just finished downloading the DVD with bittorrent. Ended up with two iso files (1 of 2 and 2 of 2). sha1sum checks out on both. This is the first time I've encountered two DVDs. Will the install politely ask for DVD #2 when it is time? ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Migrating away from Nagios
I am a pretty hardcore ZenOSS user.. We use it to monitor over 1000 devices in different fashions - using a combination of SNMP (Linux), WMI (windows) and SSH (Unix/Aix). While there is a slight learning curve to get everything working the way you want - it is, in my opinion, the most powerful open source NMS. It is a very active project with excellent community (and even commercial) support. I'd definitely suggest that you take a look at it. Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Joseph L. Casale Sent: Friday, June 18, 2010 10:32 AM To: 'centos@centos.org' Subject: [CentOS] Migrating away from Nagios I have to rebuild a new Nagios box and thought this might be a good time to migrate away. I use snmp mostly for everything but with the fork Nagios endured I wonder about putting any more effort into the project. I probably should look at OpenNMS again, but the other options I think might work are Icinga (Should be trivial to migrate) or Zenoss or maybe even Zabbix? Anyone have experience in Nagios and care to share comparisons with similar projects with strong community involvement? Also, anyone currently running OpenNMS that can comment on the learning curve and level of flexibility coming from a Nagios user? Thanks! jlc ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Cacti/snmp question
There is always ZenOSS. I would definitely take a look at ZenOSS. Very active, very powerful, nice interface, SMNP/SSH/WMI based monitoring, etc. jb -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Joseph L. Casale Sent: Wednesday, June 16, 2010 10:30 AM To: 'CentOS mailing list' Subject: Re: [CentOS] Cacti/snmp question Two, hours of attempting to get cacti to work have led me to be underimpressed with the whole project. That's odd because other than the usual php version issues I've always considered cacti to be the easiest of the graphing tools to get working - but I haven't tried the most recent versions. Last I looked at Cacti, the hack to get some plugin support didn't work for me and I didn't have the patience to waste time with it, dropped it. Munin never had zooming graphs, and needed a cgi to prevent obscene load in anything other than a trivial environment, dropped it. I have Nagios and PNP and it works well. Since your first reco to me about OpenNMS I have been intrigued, it looks like a very nice project and is very active. Ironically I do almost all my Nagios monitoring via snmp and where I can't normally use snmp, I create extends... If you are willing to hack some ugly-looking xml files that specify the oids and time intervals you can probably make opennms work for you - and you might find its other features (thresholding, notifications, etc.) useful too. Yeah, I also want to take the time to learn this package, it does look very powerful. Whit, if you are starting from scratch, I would second the reco to invest the time in OpenNMS and just learn something solid from day one. jlc ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] heartbeat package in extras trouble with 5.4
SELINUX is disabled, and I have also tried reinstalling the heartbeat related packages. No luck so far. heartbeat[8818]: 2010/05/04_22:23:37 ERROR: Cannot shmget for process status: Invalid argument This seems to be the issue. Any other ideas? Thanks From: centos-boun...@centos.org on behalf of Ryan Manikowski Sent: Tue 5/4/2010 11:23 PM To: centos@centos.org Subject: Re: [CentOS] heartbeat package in extras trouble with 5.4 On 5/4/2010 11:39 PM, Baird, Josh wrote: I just recently upgraded a box from i386 5.3 - 5.4. The box has heartbeat packages installed from extras: heartbeat-pils-2.1.3-3.el5.centos heartbeat-stonith-2.1.3-3.el5.centos heartbeat-devel-2.1.3-3.el5.centos heartbeat-2.1.3-3.el5.centos The heartbeat daemon no longer starts.. the init script reports a success, as well as the logs: May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: Enabling logging daemon May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: logfile and debug file are those specified in logd config file (default /etc/logd.cf) May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: Version 2 support: false May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: WARN: logd is enabled but logfile/debugfile is still configured in ha.cf May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: ** May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: Configuration validated. Starting heartbeat 2.1.3 May 4 22:33:10 fc-fmcln02 heartbeat: [9345]: info: heartbeat: version 2.1.3 May 4 22:33:11 fc-fmcln02 heartbeat: [9345]: info: Heartbeat generation: 1208455492 However, the daemons never actually start. When I run the daemon interactively without the init script, the following error appears: heartbeat[8818]: 2010/05/04_22:23:37 ERROR: Cannot shmget for process status: Invalid argument This may suggest that some libs on the system may have been upgraded and heartbeat is trying to use the old ones? Does anyone have any suggestions on how to get heartbeat working again? Running heartbeat on Centos 5.4 here without a problem. Just powered up my test cluster and made sure system was up-to-date using yum. Heartbeat started without a problem. Perhaps you have selinux enabled on the system? Can you try disabling selinux? This may sound like a half-hearted attempt to 'repair' the issue, but try backing up your authkeys, ha.cf and haresources on each host and try removing and reinstalling the packages. At this point you have nothing to lose since the daemons will not start. -- Ryan Manikowski ]] Devision Media Services LLC [[ www.devision.us http://www.devision.us/ r...@devision.us | 716.771.2282 ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] heartbeat package in extras trouble with 5.4
Below is a snippet of a strace: open(/usr/lib/pils/plugins/InterfaceMgr/generic.so, O_RDONLY) = 3 read(3, \177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\300\6\0\0004\0\0\0..., 512) = 512 fstat64(3, {st_mode=S_IFREG|0755, st_size=7636, ...}) = 0 mmap2(NULL, 10532, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x6ef000 mmap2(0x6f1000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1) = 0x6f1000 close(3)= 0 shmget(IPC_PRIVATE, 7816, 0600) = -1 EINVAL (Invalid argument) time(NULL) = 1273067316 open(/etc/localtime, O_RDONLY)= 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=3543, ...}) = 0 fstat64(3, {st_mode=S_IFREG|0644, st_size=3543, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7f4f000 read(3, TZif2\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\5\0\0\0\5\0\0\0\0..., 4096) = 3543 close(3)= 0 munmap(0xb7f4f000, 4096)= 0 write(2, heartbeat[28218]: 2010/05/05_08:..., 38) = 38 write(2, ERROR: Cannot shmget for process..., 58) = 58 It looks like it's trying to find an existing shared memory segment? From sysctl.conf: # Controls the maximum size of a message, in bytes kernel.msgmnb = 65536 # Controls the default maxmimum size of a mesage queue kernel.msgmax = 65536 # Controls the maximum shared segment size, in bytes kernel.shmmax = 68719476736 # Controls the maximum number of shared memory segments, in pages kernel.shmall = 4294967296 $ ipcs -- Shared Memory Segments keyshmid owner perms bytes nattch status -- Semaphore Arrays keysemid owner perms nsems 0x 2424832root 6001 0x 2719745root 6001 0x 2752514root 6001 0x 2785283root 6001 0x 2818052root 6001 0x 2850821root 6001 0x 2883590root 6001 0x 2916359root 6001 0x 3506184root 6001 0x01fe101f 3014665root 6001 0x 3407882root 6661 0x 3080203root 6001 0x 3112972root 6001 0x 3145741root 6001 0x 3178510root 6001 0x 3211279root 6001 0x 3244048root 6001 0x 3276817root 6001 0x 3309586root 6001 0x 3342355root 6001 0x 3440660root 6001 0x 3473429root 6001 -- Message Queues keymsqid owner perms used-bytes messages I'm stuck! Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Dominik Zyla Sent: Wednesday, May 05, 2010 8:10 AM To: centos@centos.org Subject: Re: [CentOS] heartbeat package in extras trouble with 5.4 On Wed, May 05, 2010 at 07:59:16AM -0500, Baird, Josh wrote: SELINUX is disabled, and I have also tried reinstalling the heartbeat related packages. No luck so far. heartbeat[8818]: 2010/05/04_22:23:37 ERROR: Cannot shmget for process status: Invalid argument This seems to be the issue. Any other ideas? Try to strace heartbeat process and check errno from shmget() and compare it against shmget(2) `ERRORS' section. Maybe you need to set some sysctls. -- Dominik Zyla ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] heartbeat package in extras trouble with 5.4
Ok, so for some reason, I had shmax set to 64GB. Prior to 5.4, I'm guessing that i386 just ignored this absurd value, but now, it forces the value to be 0: r...@fc-fmcln02:/var/log$ ipcs -l -- Shared Memory Limits max number of segments = 4096 max seg size (kbytes) = 0 max total shared memory (kbytes) = 0 min seg size (bytes) = 1 I set kernal.shnmax back to it's default value and this seemed to fix the heartbeat issues with allocating shared memory segments. Thanks, Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Baird, Josh Sent: Wednesday, May 05, 2010 9:39 AM To: CentOS mailing list Subject: Re: [CentOS] heartbeat package in extras trouble with 5.4 Below is a snippet of a strace: open(/usr/lib/pils/plugins/InterfaceMgr/generic.so, O_RDONLY) = 3 read(3, \177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\300\6\0\0004\0\0\0..., 512) = 512 fstat64(3, {st_mode=S_IFREG|0755, st_size=7636, ...}) = 0 mmap2(NULL, 10532, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x6ef000 mmap2(0x6f1000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1) = 0x6f1000 close(3)= 0 shmget(IPC_PRIVATE, 7816, 0600) = -1 EINVAL (Invalid argument) time(NULL) = 1273067316 open(/etc/localtime, O_RDONLY)= 3 fstat64(3, {st_mode=S_IFREG|0644, st_size=3543, ...}) = 0 fstat64(3, {st_mode=S_IFREG|0644, st_size=3543, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7f4f000 read(3, TZif2\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\5\0\0\0\5\0\0\0\0..., 4096) = 3543 close(3)= 0 munmap(0xb7f4f000, 4096)= 0 write(2, heartbeat[28218]: 2010/05/05_08:..., 38) = 38 write(2, ERROR: Cannot shmget for process..., 58) = 58 It looks like it's trying to find an existing shared memory segment? From sysctl.conf: # Controls the maximum size of a message, in bytes kernel.msgmnb = 65536 # Controls the default maxmimum size of a mesage queue kernel.msgmax = 65536 # Controls the maximum shared segment size, in bytes kernel.shmmax = 68719476736 # Controls the maximum number of shared memory segments, in pages kernel.shmall = 4294967296 $ ipcs -- Shared Memory Segments keyshmid owner perms bytes nattch status -- Semaphore Arrays keysemid owner perms nsems 0x 2424832root 6001 0x 2719745root 6001 0x 2752514root 6001 0x 2785283root 6001 0x 2818052root 6001 0x 2850821root 6001 0x 2883590root 6001 0x 2916359root 6001 0x 3506184root 6001 0x01fe101f 3014665root 6001 0x 3407882root 6661 0x 3080203root 6001 0x 3112972root 6001 0x 3145741root 6001 0x 3178510root 6001 0x 3211279root 6001 0x 3244048root 6001 0x 3276817root 6001 0x 3309586root 6001 0x 3342355root 6001 0x 3440660root 6001 0x 3473429root 6001 -- Message Queues keymsqid owner perms used-bytes messages I'm stuck! Josh -Original Message- From: centos-boun...@centos.org [mailto:centos-boun...@centos.org] On Behalf Of Dominik Zyla Sent: Wednesday, May 05, 2010 8:10 AM To: centos@centos.org Subject: Re: [CentOS] heartbeat package in extras trouble with 5.4 On Wed, May 05, 2010 at 07:59:16AM -0500, Baird, Josh wrote: SELINUX is disabled, and I have also tried reinstalling the heartbeat related packages. No luck so far. heartbeat[8818]: 2010/05/04_22:23:37 ERROR: Cannot shmget for process status: Invalid argument This seems to be the issue. Any other ideas? Try to strace heartbeat process and check errno from shmget() and compare it against shmget(2) `ERRORS' section. Maybe you need to set some sysctls. -- Dominik Zyla ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] heartbeat package in extras trouble with 5.4
I just recently upgraded a box from i386 5.3 - 5.4. The box has heartbeat packages installed from extras: heartbeat-pils-2.1.3-3.el5.centos heartbeat-stonith-2.1.3-3.el5.centos heartbeat-devel-2.1.3-3.el5.centos heartbeat-2.1.3-3.el5.centos The heartbeat daemon no longer starts.. the init script reports a success, as well as the logs: May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: Enabling logging daemon May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: logfile and debug file are those specified in logd config file (default /etc/logd.cf) May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: Version 2 support: false May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: WARN: logd is enabled but logfile/debugfile is still configured in ha.cf May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: ** May 4 22:33:10 fc-fmcln02 heartbeat: [9344]: info: Configuration validated. Starting heartbeat 2.1.3 May 4 22:33:10 fc-fmcln02 heartbeat: [9345]: info: heartbeat: version 2.1.3 May 4 22:33:11 fc-fmcln02 heartbeat: [9345]: info: Heartbeat generation: 1208455492 However, the daemons never actually start. When I run the daemon interactively without the init script, the following error appears: heartbeat[8818]: 2010/05/04_22:23:37 ERROR: Cannot shmget for process status: Invalid argument This may suggest that some libs on the system may have been upgraded and heartbeat is trying to use the old ones? Does anyone have any suggestions on how to get heartbeat working again? Thanks, Josh ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos