gmane.linux.keepalived.devel

Re: Solution for session persistence

Adrian Chapela — 2008-12-03T16:01:37

Adrian Chapela escribió: After some tests and after use notify_backup and notofy_master I can start and stop haproxy and monitor it with vrrp_script and track_script. Keepalived detects when is down and when is up but it doesn't do anything when the haproxy service is down, this is the finally problem. How can I do failover ? ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: Solution for session persistence

Adrian Chapela — 2008-12-03T15:34:15

Adrian Chapela escribió: Reading some on the list I think the problem is the script command, I need to write fullpath of killall command. But when it fails, how can I do the failover ? With notify_fault ? notify_backup ? ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Solution for session persistence

Adrian Chapela — 2008-12-03T09:20:11

Hello, I have a configuration of keepalived to achieve high availability and load balance of many services. One of them needs session persistence, because it is used by Blackberry devices. Blackberry has a proxy to do load balancing internet connections. One Blackberry device could have one IP in the first pettition and after login it could be another IP. For this I am thinking on haproxy and it seems to work very well. My problem is achieve high availability of haproxy service. I found this configuration to combine keepalived ( I think it could be a good solution. I am very happy with keepalived and I want to continue using it for other services like mail, DNS, etc.) with haproxy: vrrp_script chk_haproxy { # Requires keepalived-1.1.13 script "killall -0 haproxy" # cheaper than pidof interval 2 # check every 2 seconds weight 2 # add 2 points of prio if OK } vrrp_instance VI_1 { state MASTER interface eth0 vi

new keepalived release

Wolfram Schlich — 2008-11-24T15:15:01

Hi, when will the next keepalived version be released? I've read there are quite some patches to be incorporated... Thanks.

Re: keepalived VRRP part used for N+1 redundant services

2008-11-19T07:43:46

Hi, I've found the problem in the meantime. I found out that I missed a closing brace in my config file :-[ so both VRRP instances had collapsed to one. I only found this out after digging into the source code of keepalived :-\ Thanks to all that tried to help me out ;-) Wouter. On Tue, 18 Nov 2008 17:16:12 +0000, "Graeme Fowler" graemef.net> said: ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: keepalived VRRP part used for N+1redundantservices

Martin Barry — 2008-11-18T18:45:00

$quoted_author = "Graeme Fowler" ; This isn't quite the setup he wants. Server 1 = Master A Server 2 = Master B Server 3 = Backup A, Backup B That is still probably the case. We'd need to see the master configs. cheers marty

Re: keepalived_healthcheckers flooding syslog

Jack Neely — 2008-11-18T18:18:41

Thanks for the responce. I did find that the RHEL 4 platform logged a failing healthcheck multiple times...but there were 15 minutes to an hour before it would log the same failure again. So...doesn't make much sense here either. Fortunately, I've been able to fix the failures by removing people's old mess. :-) Jack

Re: keepalived VRRP part used for N+1redundantservices

Graeme Fowler — 2008-11-18T17:16:12

Hi On Thu, 2008-11-13 at 14:00 +0100, wouterkalived< at >fastmail.net wrote: Yep, pretty easy to do. There's no need to shout :) This is perfectly normal - I have a couple running rather like this, master/backup on one director and backup/master on the other. If it complains about that, your configs are banjaxed. Make sure you have the same IP addresses in the same vrrp_instances, with the same virtual_router_id (VRID). It looks from here - although you didn't give the master config - that you have ever-so-slightly different config for each VRID, so it's complaining. At work we've got a boatload of VRRP instances spread across multiple pairs (or more) of servers, all on the same /24 subnet, and they all get along fine. Graeme ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source even

Re: keepalived_healthcheckers flooding syslog

Graeme Fowler — 2008-11-18T17:10:10

Possibly the way syslog is configured - it can "eat" messages and print occasional "last message repeated N times" entries in order to save space. Other than that, no idea at all - if it's the same version, it doesn't make a heap of sense to me. Graeme ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

keepalived VRRP part used for N+1 redundantservices

2008-11-13T13:00:02

Hello, I would like to use just the VRRP part of keepalived to setup an N+1 redundancy scheme for a service that runs on the VRRP hosts. Just to give a simple example: - 2 active web servers, one on IP 1.2.3.4 and on IP 1.2.3.5 - 1 backup web server, that can take over one (or both) of the IP addresses given above (and will also take over the WEB service) This is how the configuration file would look like on the backup server: ---- vrrp_instance VI_10 { state BACKUP interface eth0 virtual_router_id 10 priority 50 authentication { auth_type PASS auth_pass password } virtual_ipaddress { 1.2.3.4/24 brd 10.0.11.255 dev eth0 } vrrp_instance VI_11 { state BACKUP interface eth0 virtual_router_id 11 priority 40 authentication { auth_type PASS auth_pass password } virtual_ipaddress { 1.2.3.5/23 brd 10.0.11.255 dev eth0 } ---- So BOTH VRRP INSTANCES WOULD BE RUNNING ON THE SAME ETHERNET INTERFACE (eth0). If I try to configure it like this, keepalived complains about a bogus VRRP packet,

keepalived_healthcheckers flooding syslog

Jack Neely — 2008-11-11T22:33:19

Folks, I'm migrating keepalived 1.1.15 from old servers running RHEL 4 to new servers running RHEL 5. I've had no problems moving VIPs between the different platforms, in fact its worked quite smoothly. However, the healthchecker thread seems to flood syslog on the RHEL 5 servers. Every delay_loop interval I get a log from each of the three RS that happen to be down: Nov 11 17:29:36 lvs00 Keepalived_healthcheckers: Timeout connect, timeout server [xxx.xxx.xxx.aaa:7777]. Nov 11 17:29:36 lvs00 Keepalived_healthcheckers: Timeout connect, timeout server [xxx.xxx.xxx.bbb:80]. Nov 11 17:29:36 lvs00 Keepalived_healthcheckers: Timeout connect, timeout server [xxx.xxx.xxx.ccc:80]. On the RHEL 4 platform I only got one syslog message when a host failed its healthcheck and it was silent until the RS came back online. Any idea what the difference may be? Jack Neely

Re: [lvs-users] Https sorry server problems

Adrian Chapela — 2008-11-10T10:30:21

Graeme Fowler escribió: I am re-reading this again: http://www.austintek.com/LVS/LVS-HOWTO/HOWTO/LVS-HOWTO.localnode.html I am sure now that my problem was my lighthttpd configuration. This is a paragraph of the HOWTO: "...You then setup the service to listen to the VIP on the director..." Thank you for open my eyes! ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [lvs-users] Https sorry server problems

Adrian Chapela — 2008-11-10T10:01:03

Graeme Fowler escribió: No, my server is listen only on 127.0.0.1 and in the real IP of the director. Could be this the problem ? In http my server is listen on VIP also. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [lvs-users] Https sorry server problems

Graeme Fowler — 2008-11-10T09:42:29

There's something here that I can't quite put my finger on. I can almost smell the solution but it's proving tantalisingly out of reach! I've just tested this using a very basic configuration, with a single real_server that doesn't have a listener on port 443. My config looked like this: virtual_server 192.168.1.201 443 { delay_loop 6 lb_algo rr lb_kind DR nat_mask 255.255.255.0 persistence_timeout 50 protocol TCP sorry_server 127.0.0.1 443 real_server 192.168.2.215 443 { weight 100 TCP_CHECK { connect_timeout 3 } } } But the notable thing here is that the local Apache version on the director is also listening on the VIP on port 443, too. Is yours? Graeme ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source e

keepalived - virtual_routes allow/enable tablexyz option

Pieter Smit — 2008-11-07T19:46:30

------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/_______________________________________________ Keepalived-devel mailing list Keepalived-devel< at >lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/keepalived-devel

Re: [lvs-users] Https sorry server problems

Adrian Chapela — 2008-11-07T12:17:32

Graeme Fowler escribió: Yes my server is configured to listen in 127.0.0.1:443 but when all of my real servers are down, I can't connect with nothing (because all of my servers are down and I can't connect wiht my sorry server). Yes, I have certificate, key correct. I use lighthttpd for this. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [lvs-users] Https sorry server problems

Graeme Fowler — 2008-11-07T12:12:28

OK, all sane. However, SSL sites often don't bind to wildcard addresses (unlike those on port 80). You need to configure your webserver to listen to 127.0.0.1:443 with an appropriate SSL certificate. Note also that you may get certificate warnings unless you have configured the sorry server with the right certificate/key pair for the virtual host it's handling. Regards Graeme ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [lvs-users] Https sorry server problems

Adrian Chapela — 2008-11-07T11:53:36

Graeme Fowler escribió: Yes, it is the config: virtual_server group WEB_SSL { delay_loop 6 lb_algo wrr lb_kind DR persistence_timeout 3600 protocol TCP sorry_server 127.0.0.1 443 real_server 192.168.0.1 443 { weight 90 MISC_CHECK { misc_path "/usr/bin/check_http -S -H 192.168.0.1 -p 443 -t 8" misc_timeout 6 } } real_server 192.168.0.2 443 { weight 5 MISC_CHECK { misc_path "/usr/bin/check_http -S -H 192.168.0.2 -p 443 -t 8" misc_timeout 6 } } real_server 192.168.0.3 443 { weight 5 MISC_CHECK { misc_path "/usr/bin/check_http -S -H 192.168.0.3 -p 443 -t 8" misc_timeout 6 } } } I use direct connection. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based appl

Re: [lvs-users] Https sorry server problems

Graeme Fowler — 2008-11-07T11:01:09

Hi Adrian On Fri, 2008-11-07 at 11:06 +0100, Adrian Chapela wrote: You should really be asking this on the keepalived list, so I've copied that one in. Can you show us the config section for the virtual_server, please? Graeme ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: is keepalived still under development ?

Robin Garner — 2008-11-02T23:39:09

There are two issues we have with keepalived at the moment. The first is syntax checking - currently if you have an error in the configuration file, most of the time keepalived silently ignores the error and just doesn't do what you intended. I'm currently attempting to write a "--checkconfig" option that will report syntax errors and return a good/bad status code. We're using cfengine to manage the configuration file, and being able to write a script that says (essentially) if config_ok; then reload keepalived; else send mail error; fi will make life a lot easier. The second issue is some status reporting. For example, using 'ip addr show' to check the status of the VRRP servers is OK, but sometimes we've found that both servers have all the virtual addresses defined, and it would be nice to be able to see what keepalived thinks is going on (and of course to fix the underlying bug). Robin ------------------------------------------------------------------------- This SF.Net email is spo

Re: is keepalived still under development ?

Alexandre Cassen — 2008-10-30T21:47:30

'lo, On Thu, 30 Oct 2008, J?r?me Loyet wrote: no CVS, just send the diff -urN, will merge then -- Alexandre ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/