Re: [CentOS] scripting CPAN installs
I think this could be done by cpan o conf prerequisites_policy follow cpan o conf commit and then set env PERL_MM_USE_DEFAULT=1 cpan or env PERL_MM_USE_DEFAULT=1 perl -MCPAN -e shell Tomas Mon, Apr 19, 2010 ve 04:24:21PM -0500, Les Mikesell napsal: On 4/19/2010 4:15 PM, Kahlil Hodgson wrote: On 04/17/2010 12:21 AM, Alan McKay wrote: Now my only problem seems to be that cpan does not allow me to specify yes to this question when it asks. e.g. like the -y option to yum Unsatisfied dependencies detected during [A/AN/ANDK/CPAN-1.9402.tar.gz] - Test::Harness Shall I follow them and prepend them to the queue of modules we are processing right now? [yes] Any way to do that? Try piping the output of the 'yes' command to whatever script is asking the question. man yes for more details ;-) I thought that was one of the things you could set to default to 'yes' when you initially configure the cpan module, but it's been a long time since I did that. -- Les Mikesell lesmikes...@gmail.com ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] scripting CPAN installs
Hi, i don't know exactly if this would be usable for yourself, but very nice feature for our purposes is autobundling, see for example: http://search.cpan.org/~andk/CPAN-1.9402/lib/CPAN.pm#POPULATE_AN_INSTALLATION_WITH_LOTS_OF_MODULES or http://www.developertutorials.com/tutorials/cgi-perl/automate-perl-module-deployment-050426/page4.html As for RPMs, it could be sometimes problem with versions. RPMs aren't often the most recent versions of modules and resolving RPM dependencies could sometimes overwrite the most uptodate versions (which you need) by slightly older which may not be applicable. But this really depends... Tomas Thu, Apr 15, 2010 ve 04:23:32PM -0400, Alan McKay napsal: Hey folks, Maybe there is a Perl/CPAN list that is a better place to ask this? If so, maybe someone can point me to it. Anyway, I want to be able to script the installation of a bunch of CPAN modules, and the first basic problem I am coming up against is that the cpan command seems to always return 0 regardless of whether or not the install completed. Google does not bring up a whole lot of help for me here, but I have to think this problem has already been solved. Thanks for any guidance you can give. -Alan -- ?Don't eat anything you've ever seen advertised on TV? - Michael Pollan, author of In Defense of Food ___ 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] scripting CPAN installs
Sorry, sent too soon. As for your problem with not knowing if the module was succesfully installed... i went around by trying use/require and printing version after each install. Tomas Fri, Apr 16, 2010 ve 03:26:16PM +0200, Tomas Ruprich napsal: Hi, i don't know exactly if this would be usable for yourself, but very nice feature for our purposes is autobundling, see for example: http://search.cpan.org/~andk/CPAN-1.9402/lib/CPAN.pm#POPULATE_AN_INSTALLATION_WITH_LOTS_OF_MODULES or http://www.developertutorials.com/tutorials/cgi-perl/automate-perl-module-deployment-050426/page4.html As for RPMs, it could be sometimes problem with versions. RPMs aren't often the most recent versions of modules and resolving RPM dependencies could sometimes overwrite the most uptodate versions (which you need) by slightly older which may not be applicable. But this really depends... Tomas Thu, Apr 15, 2010 ve 04:23:32PM -0400, Alan McKay napsal: Hey folks, Maybe there is a Perl/CPAN list that is a better place to ask this? If so, maybe someone can point me to it. Anyway, I want to be able to script the installation of a bunch of CPAN modules, and the first basic problem I am coming up against is that the cpan command seems to always return 0 regardless of whether or not the install completed. Google does not bring up a whole lot of help for me here, but I have to think this problem has already been solved. Thanks for any guidance you can give. -Alan -- ?Don't eat anything you've ever seen advertised on TV? - Michael Pollan, author of In Defense of Food ___ 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 mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to log separate files or directories for centralizing SysLog server ?
Hi James, i think much better for syslog server usage is syslog-ng. It has bit more difficult configuration for such a small network, but it brings much more functionalities. the simple ruleset for your needs could look like: source s_sys { unix-stream(/dev/log); internal(); }; source s_net { udp(); }; destination d_net { file(/var/log/network/$HOST/$YEAR.$MONTH.log owner(root) group(root) perm(0600) dir_perm(0700) create_dirs(yes)); }; log { source(s_net); destination(d_net); }; log { source(s_sys); destination(d_net); }; Hope it helps, Tomas Mon, Apr 12, 2010 ve 03:49:53PM +0800, James Corteciano napsal: Hi All, I have three server. server0 is centralized logging server, server1 and server2 are remote client servers. How can I properly configure the syslog in server0 to log the two servers in different separated files/directories. Example, server1 will be logged at /var/log/syslog/server1.log of server0 and /var/log/syslog/server2.log for server2 as well. Is there need of little tweaking for syslog.conf ? Thanks. James ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to log separate files or directories for centralizing SysLog server ?
Not that rules, but definetely it's possible with rsyslog. http://www.rsyslog.com/Article60.phtml Tomas Mon, Apr 12, 2010 ve 04:12:39PM +0800, James Corteciano napsal: Hi Tomas, I can't use syslog-ng because it's not included in RHEL package in DVD and company policy not to use non-rpm. I can use rsyslog and found it's the same config to syslog. Can I apply that rules in rsyslog? Thanks. James On Mon, Apr 12, 2010 at 3:57 PM, Tomas Ruprich rupr...@uikt.mendelu.czwrote: Hi James, i think much better for syslog server usage is syslog-ng. It has bit more difficult configuration for such a small network, but it brings much more functionalities. the simple ruleset for your needs could look like: source s_sys { unix-stream(/dev/log); internal(); }; source s_net { udp(); }; destination d_net { file(/var/log/network/$HOST/$YEAR.$MONTH.log owner(root) group(root) perm(0600) dir_perm(0700) create_dirs(yes)); }; log { source(s_net); destination(d_net); }; log { source(s_sys); destination(d_net); }; Hope it helps, Tomas Mon, Apr 12, 2010 ve 03:49:53PM +0800, James Corteciano napsal: Hi All, I have three server. server0 is centralized logging server, server1 and server2 are remote client servers. How can I properly configure the syslog in server0 to log the two servers in different separated files/directories. Example, server1 will be logged at /var/log/syslog/server1.log of server0 and /var/log/syslog/server2.log for server2 as well. Is there need of little tweaking for syslog.conf ? Thanks. James ___ 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 mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos -- S pozdravem Tomáš Ruprich systémový administrátor Ústav pro informační systém Mendelova univerzita v Brně Zemědělská 1 / 613 00 Brno telefon 545 132 885 rupr...@uikt.mendelu.cz www.mendelu.cz [prostor pro logo] ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] iptables
Wed, Sep 16, 2009 ve 03:31:43PM +0800, CentOS List napsal: snip *nat # Manipulate nat table :PREROUTING ACCEPT [0:0] :POSTROUTING ACCEPT [0:0] :OUTPUT ACCEPT [0:0] -A PREROUTING -t nat -p tcp --dport 26 -j REDIRECT --to-port 25 COMMIT Thanks Christopher, I had updated my rules to *nat :OUTPUT ACCEPT [0:0] :PREROUTING ACCEPT [0:0] :POSTROUTING ACCEPT [0:0] -A PREROUTING -t nat -p tcp --dport 26 -j REDIRECT --to-port 25 COMMIT # Completed When I did a restart, there is an error. Flushing firewall rules: [ OK ] Setting chains to policy ACCEPT: nat mangle filter [ OK ] Unloading iptables modules:[ OK ] Applying iptables firewall rules: iptables-restore v1.3.5: Line 52 seems to have a -t table option. Error occurred at line: 52 Try `iptables-restore -h' or 'iptables-restore --help' for more information. [FAILED] Line 52 is -A PREROUTING -t nat -p tcp --dport 26 -j REDIRECT --to-port 25 Regards Never (if you aren't guru) edit your iptables by directly editing your config. Always use iptables binary. It is said there... -t option shouldn't be on that line. The line should be only -A PREROUTING -p tcp --dport 26 -j REDIRECT --to-port 25 because you're already in the *nat section of the config. But again, you should enter this command instead of editing the config: iptables -A PREROUTING -t nat -p tcp --dport 26 -j REDIRECT --to-port 25 regards Tomas ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] logical volume over NFS permissions problem
Hi, it is exactly as you're writing, thanks for help. I've been trying to solve it with NFSv4 for a short while, but it behaved the same way. So at last, i configured autofs, which is mounting all the logical volume mountpoints itself. Just for the others, here's the way i configured autofs: [r...@client]# echo /net /etc/auto.escience /etc/auto.master [r...@client]# echo * -rw,intr,soft,timeo=300,tcp,rsize=32768,wsize=32768 nfs_server:/export/escience/ /etc/auto.escience [r...@client]# service autofs restart [r...@client]# touch /net/name_of_the_share Thanks again Filipe. Tomas Thu, Jun 25, 2009 ve 09:18:11AM -0400, Filipe Brandenburger napsal: Hi, 2009/6/25 Tomas Ruprich rupr...@uikt.mendelu.cz: i've problem with mounting logical volumes over NFS. You are mounting /export/escience and you want to access /export/escience/vv_25 which is another filesystem. NFS (v2 and v3) does not work that way, if you export and mount /export/escience, you will get only the files in that filesystem and not in filesystems mounted below it. It does not traverse mount points. To achieve what you want, you have to export and mount /export/escience/vv_25 explicitely. Or you can investigate NFSv4, which AFAIR works the way you are trying to use, but it has many differences from NFSv3 and there is some learning curve involved. HTH, Filipe ___ 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] logical volume over NFS permissions problem
Hi all, i've problem with mounting logical volumes over NFS. On NFS server, i have following logical volume [r...@nfs_server ~]# lvdisplay --- Logical volume --- LV Name/dev/escience/vv_25 VGnfs_servernfs_servernfs_servernfs_servernfs_server escience LV UUIDiMhNq7-iC7L-VTDO-Xcwv-a6yv-sEAD-EZWASV LV Write Accessread/write LV Status available # open 0 LV Size5.00 GB Current LE 1280 Segments 1 Allocation inherit Read ahead sectors auto - currently set to 256 Block device 253:2 This logical volume is mounted on NFS server, the group information is taken from LDAP server. [r...@nfs_server ~]# mount | grep vv_25 /dev/mapper/escience-vv_25 on /export/escience/vv_25 type ext3 (rw) [r...@nfs_server ~]# ls -ld /export/escience/vv_25 drwxrwx--- 3 root vv153 4096 Jun 24 16:04 /export/escience/vv_25 But when i mount /export/escience on the nfs client, i'm missing all the permissions and ownership informations: [r...@client ~]# mount.nfs nfs_server:/export/escience/ /mnt/escience [r...@client ~]# ls -ld /mnt/escience/vv_25 drwxr-xr-x 2 root root 4096 2009-06-24 16:04 vv_25 The setting of NFS exports is following: [r...@nfs_server ~]# cat /etc/exports /export/escience client(rw,sync,no_root_squash) [r...@nfs_server ~]# exportfs -v /export/escience client(rw,wdelay,no_root_squash,no_subtree_check,anonuid=65534,anongid=65534) Server is running latest CentOS 5.3 [r...@nfs_server ~]# cat /etc/redhat-release CentOS release 5.3 (Final) [r...@nfs_server ~]# uname -a Linux nfs_server 2.6.18-128.1.14.el5 #1 SMP Wed Jun 17 06:38:05 EDT 2009 x86_64 x86_64 x86_64 GNU/Linux No denials from selinux, client's IP is completely opened on firewall. Is there some possibility to debug the NFS mount process? I don't see nothing more then verbose mode in man, which doesn't provide much informations :( Or is it some configuration issue? thanks for help, Tomáš Ruprich rupr...@uikt.mendelu.cz DCD IICT MUAF Brno www.mendelu.cz, is.mendelu.cz to...@jabber.cz ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] HA software advice
yum install heartbeat Very robust, very reliable, easy to configure, easy to use :-) We use it for almost every critical server for about 3 years without any problem. http://www.linux-ha.org/ Tomáš Ruprich [EMAIL PROTECTED] DCD IICT MUAF Brno www.mendelu.cz, is.mendelu.cz tel.: +420 545 132 885, +420 602 127 744 [EMAIL PROTECTED] Wed, Jan 23, 2008 ve 03:51:13PM -0500, Scott McClanahan napsal: We are in need of some very basic software that will give us the ability to swing an ip address from one host to another during controlled maintenance or host failure. For now the IP address will be the only resource that is shared and there will never be a need for shared storage. Eventually we may want to monitor processes for health (probably just read in pid file) and it would be great if the recommended software had this ability as well but not completely necessary. The process monitoring part is just me trying to think ahead but it definitely wouldn't be needed in the near future. Ideally, this should be light weight user land code that hopefully already comes easily with the distribution :) Any suggestions? Thanks. ___ 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] HA software advice
Wed, Jan 23, 2008 ve 04:24:43PM -0500, Scott McClanahan napsal: On Wed, 2008-01-23 at 22:20 +0100, Tomas Ruprich wrote: yum install heartbeat Very robust, very reliable, easy to configure, easy to use :-) We use it for almost every critical server for about 3 years without any problem. http://www.linux-ha.org/ Tomáš Ruprich [EMAIL PROTECTED] DCD IICT MUAF Brno www.mendelu.cz, is.mendelu.cz tel.: +420 545 132 885, +420 602 127 744 [EMAIL PROTECTED] I had originally considered heartbeat but just didn't know if version 2 was overkill. It's a pretty capable package and I just didn't want to get in over my head for this project. Is version 2 still pretty simple if you want it to be? Well, exactly as you say... if you want it to be, it can be still pretty simple :) I don't know any software which could be simplier and still so reliable for mission critical usage. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] Directory index forbidden by rule: /var/www/html/
Directory index is a feature of apache, which: - if there is a file named in DirectoryIndex directive (usually index.html) in the requested directory, then show it (i think this is made by mod_dir) - if not, then create a list of files, if you have mod_autoindex loaded and Options +Indexes in directory configuration (I recommend http://httpd.apache.org/docs/2.2/mod/mod_dir.html and http://httpd.apache.org/docs/2.2/mod/mod_autoindex.html) I believe, this comes because your /var/www/html contains no index mentioned in DirectoryIndex rule. What is the content of /var/www/html? best regards --- Tomáš Ruprich [EMAIL PROTECTED] DCD IICT MUAF Brno tel.: +420 545 132 885, +420 602 127 744 --- Tue, Oct 09, 2007 ve 08:34:21AM -0700, Rogelio Bastardo napsal: I'm looking to troubleshoot this error when I run tail /etc/httpd/logs/error_log [Tue Oct 09 07:22:59 2007] [error] [client 127.0.0.1] Directory index forbidden by rule: /var/www/html/ (I get this when I run Nagios' ./check_http -H 127.0.0.1.) All I did was install CentOS and then modify httpd.conf to include ServerName 192.168.1.10:80 and also run htpasswd -c /etc/nagios/htpasswd.users nagiosadmin. Is this a permissions issue on my /var/www/html/ folder? I tried to get the Apache page to serve properly, and it's not doing that. And when I try from another computer, I don't see anything in the logs, which made me think it was an IPTABLES issue (which it doesn't appear to be). Chain RH-Firewall-1-INPUT (2 references) target prot opt source destination ACCEPT all -- anywhere anywhere ACCEPT icmp -- anywhere anywhereicmp any ACCEPT ipv6-crypt-- anywhere anywhere ACCEPT ipv6-auth-- anywhere anywhere ACCEPT udp -- anywhere 224.0.0.251 udp dpt:5353 ACCEPT udp -- anywhere anywhereudp dpt:ipp ACCEPT all -- anywhere anywherestate RELATED,ESTAB LISHED ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:s sh ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:h ttp ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:f tp ACCEPT tcp -- anywhere anywherestate NEW tcp dpt:s mtp REJECT all -- anywhere anywherereject-with icmp-ho st-prohibited ___ 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] strange chgrp behavior
Greetings, i had a following problem, which on one side i have already solved, but on other side i don't thing it should behave this way and wish somebody to tell if it is feature or bug ;-) [EMAIL PROTECTED] home]# chown user user/ [EMAIL PROTECTED] home]# ls -ld /home/user drwxr-xr-x 66 user staff 4096 /home/user [EMAIL PROTECTED] home]# chgrp -R staff user/ [EMAIL PROTECTED] home]# ls -ld user/ drwxr-xr-x 66 root staff 4096 user/ ^^ I've been working on it hole afternoon, playing with user permissions, moving home directories to others, debugging selinux and other things, but at last, it was caused by [EMAIL PROTECTED] home]# ls -la user/.openoffice/user/work lrwxrwxrwx 1 root root 13 Aug 28 14:35 user/.openoffice/user/work - /home/user When i remove this symlink, everything works fine. I tried this on CentOS 4.4 and 4.5 with coreutils-5.2.1-31.4 and coreutils-5.2.1-31.6 and both are doing this. CentOS 5 with coreutils-5.97-12.1.el5 is already fine. I've been searching through coreutils gnu.org archives, but found nothing (maybe i'm not so experienced seeker ;-)) Thanks --- Tomáš Ruprich [EMAIL PROTECTED] DCD IICT MUAF Brno tel.: +420 545 132 885, +420 602 127 744 --- ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] strange chgrp behavior
Thu, Aug 30, 2007 ve 06:05:49PM +0200, Tomas Ruprich napsal: Greetings, i had a following problem, which on one side i have already solved, but on other side i don't thing it should behave this way and wish somebody to tell if it is feature or bug ;-) [EMAIL PROTECTED] home]# chown user user/ [EMAIL PROTECTED] home]# ls -ld /home/user drwxr-xr-x 66 user staff 4096 /home/user [EMAIL PROTECTED] home]# chgrp -R staff user/ [EMAIL PROTECTED] home]# ls -ld user/ drwxr-xr-x 66 root staff 4096 user/ ^^ just to be accurate, this doesn't happen when using --no-dereference and it is not changing permissions to root, but to user who is the owner of the link, so: [EMAIL PROTECTED] home]# ls -la user/.openoffice/user/work lrwxrwxrwx 1 nobody root 13 Aug 28 14:35 user/.openoffice/user/work - /home/user [EMAIL PROTECTED] home]# ls -ld /home/user drwxr-xr-x 66 user staff 4096 /home/user [EMAIL PROTECTED] home]# chgrp -R staff user/ [EMAIL PROTECTED] home]# ls -ld user/ drwxr-xr-x 66 nobody staff 4096 user/ --- Tomáš Ruprich [EMAIL PROTECTED] DCD IICT MUAF Brno tel.: +420 545 132 885, +420 602 127 744 --- ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] GFS availability in Centos5
Hi, try http://www.centos.org/docs/5/html/Global_File_System/ and http://www.centos.org/docs/5/html/Global_File_System/s1-sysreq-rhcs.html regards - Tomáš Ruprich [EMAIL PROTECTED] OKV UIKT MUAF Brno tel.: +420 545 132 885, +420 602 127 744 Tue, Aug 21, 2007 ve 03:05:32PM +0200, Joachim Backes napsal: Hi, does Centos-5 support the GFS filesystem and the RedHat cluster suite? I did not find anything in the FAQs. Regards Joachim Backes [EMAIL PROTECTED] University of Kaiserslautern,Computer Center [RHRK], Systems and Operations, High Performance Computing, D-67653 Kaiserslautern, PO Box 3049, Germany -- Phone: +49-631-205-2438, FAX: +49-631-205-3056 http://hlrwm.rhrk.uni-kl.de/home/staff/backes.html ___ 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] Diskless client from system-config-netboot doesn't boot.
Tue, Aug 21, 2007 ve 02:42:33PM -0400, David Mackintosh napsal: On Mon, Aug 20, 2007 at 10:14:51AM -0400, R P Herrold wrote: On Mon, 20 Aug 2007, Sophana wrote: David Mackintosh a écrit : I've followed a set of instructions I found on http://www.linuxtopia.org/online_books/centos_linux_guides/centos_enterprise_linux_sysadmin_guide/ch-diskless.htmli which describes using system-config-netboot to set up PXE booting. I just can say that I had the same problem with centos 4.5. Seems that centos is a little buggy on this. and which bug is that again in the centos bug tracker? I personally didn't enter a bug, as my research on the subject showed a lot of discussion in the Fedora distro as to what this tool was really supposed to be doing. It isn't included in the upstream v5 release, but is due to get re-included at a later time once things are sorted out. I think this is rather like the Xen offering -- the bare bones of a good/useful idea that needs more work before it's ready for general use. I'm sure it will be usable in the future, just not today. I have access to, but have not tried, the upstream product. If I get a chance I will and if it functions differently than the CentOS offering I will enter a bug; any other problems are more likely upstream problems. Hi, i'm not familiar with precisely this solution, but i've been solving something similar few months ago... I think the best thing you should do when searching for the reason is going step after step after the mount of pivot root. There must be some script on the tftp server, where the mount of the pivot root and snapshot is executed... so find it and then start shell right after it. This gives you enough space for finding what is happening and trying how to fix or find the bug (if it is a bug ;-)) best regards --- Tomáš Ruprich [EMAIL PROTECTED] DCD IICT MUAF Brno tel.: +420 545 132 885, +420 602 127 744 --- ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] is there kiosk mode for firefox?
I don't know IE kiosk mode, byt try r-kiosk (https://addons.mozilla.org/en-US/firefox/addon/1659). Opera has quite nice kiosk mode too ;-) Tom Sun, Jul 22, 2007 ve 09:24:48PM +0800, Fung napsal: Is there kiosk mode for firefox in linux, like the IE kiosk mode in windows? ___ 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