Re: [Pacemaker] pingd
On Thu, Sep 02, 2010 at 09:33:59PM +0200, Andrew Beekhof wrote: > On Thu, Sep 2, 2010 at 4:05 PM, Lars Ellenberg > wrote: > > On Thu, Sep 02, 2010 at 11:00:12AM +0200, Bernd Schubert wrote: > >> On Thursday, September 02, 2010, Andrew Beekhof wrote: > >> > On Wed, Sep 1, 2010 at 11:59 AM, Bernd Schubert > >> > > My proposal is to rip out all network code out of pingd and to add > >> > > slightly modified files from 'iputils'. > >> > > >> > Close, but thats not portable. > >> > Instead use ocf:pacemaker:ping which goes a step further and ditches > >> > the daemon piece altogether. > >> > >> Hmm, we are already using that for now temporarily. But I don't think the > >> ping > >> RA is suitable for larger clusters. The ping script RA runs everything > >> serially and only in intervals when called by lrmd. Now lets assume we > >> have a > >> 20 node cluster. > >> > >> nodes = 20 > >> timeout = 2 > >> attempts = 2 > >> > >> Makes 80s for a single run with default already rather small timeouts, > >> which > >> is IMHO a bit large. And with a shell script I don't see a way to improve > >> that. While we could send the pings in parallel, I have no idea how to lock > >> the variable of active nodes (active=`expr $active + 1`). I don't think > >> that > >> the simple sh or even bash have a semaphore or mutex lock. So IMHO, we > >> need a > >> language that supports that, rewriting the pingd RA is one choice, > >> rewriting > >> the ping RA into python is another. > > > > how about an fping RA ? > > active=$(fping -a -i 5 -t 250 -B1 -r1 $host_list 2>/dev/null | wc -l) > > > > terminates in about 3 seconds for a hostlist of 100 (on the LAN, 29 of > > which are alive). > > Happy to add if someone writes it :-) I thought so ;-) Additional note to whomever is going to: With fping you can get fancy about "better connectivity", you are not limited to the measure "number of nodes responding". You could also use the statistics on packet loss and rtt provided on stderr for -c or -C mode (example output below, chose what you think is easier to parse), then do some scoring scheme on average or max packet loss, rtt, or whatever else makes sense to you. (If a switch starts dying, it may produce increasing packet loss first...) Or start a smokeping daemon, and use the triggers there to change pacemaker attributes. Uhm, well, thats probably no longer maintainable, though ;-) # fping -q -i 5 -t 250 -B1 -r2 -C5 -g 10.9.9.50 10.9.9.70 10.9.9.50 : 0.14 0.14 0.16 0.12 0.15 10.9.9.51 : - - - - - 10.9.9.52 : - - - - - 10.9.9.53 : 0.37 0.34 0.36 0.34 0.34 10.9.9.54 : 0.13 0.12 0.13 0.12 0.13 10.9.9.55 : 0.17 0.15 0.16 0.12 0.22 10.9.9.56 : 0.32 0.32 0.31 0.41 0.36 10.9.9.57 : 0.35 0.33 0.32 0.34 0.32 10.9.9.58 : - - - - - 10.9.9.59 : - - - - - 10.9.9.60 : - - - - - 10.9.9.61 : - - - - - 10.9.9.62 : - - - - - 10.9.9.63 : - - - - - 10.9.9.64 : - - - - - 10.9.9.65 : 1.92 0.33 0.33 0.33 0.34 10.9.9.66 : - - - - - 10.9.9.67 : - - - - - 10.9.9.68 : - - - - - 10.9.9.69 : 0.15 0.14 0.17 0.13 0.14 10.9.9.70 : - - - - - # fping -q -i 5 -t 250 -B1 -r2 -c5 -g 10.9.9.50 10.9.9.70 10.9.9.50 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.11/0.13/0.15 10.9.9.51 : xmt/rcv/%loss = 5/0/100% 10.9.9.52 : xmt/rcv/%loss = 5/0/100% 10.9.9.53 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.33/0.34/0.37 10.9.9.54 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.10/0.11/0.13 10.9.9.55 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.13/0.16/0.20 10.9.9.56 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.34/0.36/0.41 10.9.9.57 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.16/0.25/0.33 10.9.9.58 : xmt/rcv/%loss = 5/0/100% 10.9.9.59 : xmt/rcv/%loss = 5/0/100% 10.9.9.60 : xmt/rcv/%loss = 5/0/100% 10.9.9.61 : xmt/rcv/%loss = 5/0/100% 10.9.9.62 : xmt/rcv/%loss = 5/0/100% 10.9.9.63 : xmt/rcv/%loss = 5/0/100% 10.9.9.64 : xmt/rcv/%loss = 5/0/100% 10.9.9.65 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.28/0.32/0.34 10.9.9.66 : xmt/rcv/%loss = 5/0/100% 10.9.9.67 : xmt/rcv/%loss = 5/0/100% 10.9.9.68 : xmt/rcv/%loss = 5/0/100% 10.9.9.69 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.13/0.14/0.15 10.9.9.70 : xmt/rcv/%loss = 5/0/100% > >> So in fact my first proposal also only was the first step - first add > >> better > >> network code and then to make it multi-threaded - each ping host gets its > >> own > >> thread. > > > > A working pingd daemon has the additional advantage that it can ask its > > peers for their ping node count, before actually updating the attribute, > > which should help with the "dampen race". > > That happens at the attrd level in both cases. pingd adds nothing here. I thought pingd did the dampening itself, even communicated with its peer pingd's, and there was no more dampening in attrd involved after that. But If you say so. I never looked at pingd too closely. > >> PS: (*) As you insist ;) on quorum with n/2 + 1 nodes, we use ping as > >> replacement. We simply cannot fulfill n/2 + 1, as controller failure takes > >> down 50% of the systems (virtual machines) and the systems (
Re: [Pacemaker] pingd
On Friday, September 03, 2010, Lars Ellenberg wrote: > > > how about an fping RA ? > > > active=$(fping -a -i 5 -t 250 -B1 -r1 $host_list 2>/dev/null | wc -l) > > > > > > terminates in about 3 seconds for a hostlist of 100 (on the LAN, 29 of > > > which are alive). > > > > Happy to add if someone writes it :-) > > I thought so ;-) > Additional note to whomever is going to: > > With fping you can get fancy about "better connectivity", > you are not limited to the measure "number of nodes responding". I think for the beginning, just the basic feature should be sufficient. Actually I thought about to add an option to the existing ping RA to let the user choose between ping and fping, it would default to ping. I will do that mid of next week. > You could also use the statistics on packet loss and rtt provided on > stderr for -c or -C mode (example output below, chose what you think is > easier to parse), then do some scoring scheme on average or max packet > loss, rtt, or whatever else makes sense to you. > (If a switch starts dying, it may produce increasing packet loss first...) That will require quite parsing, which I'm not comfortable with in a shell script. I have no objections to later on add fping RA written in perl or python. [...] > > > >> PS: (*) As you insist ;) on quorum with n/2 + 1 nodes, we use ping as > > >> replacement. We simply cannot fulfill n/2 + 1, as controller failure > > >> takes down 50% of the systems (virtual machines) and the systems > > >> (VMs) of the 2nd controller are then supposed to take over failed > > >> services. I see that n/2 + 1 is optimal and also required for a few > > >> nodes. But if you have a larger set of system (e.g. minimum 6 with > > >> the VM systems I have in my mind) n/2 + 1 is sufficient, IMHO. > > > > > > You meant to say you consider == n/2 sufficient, instead of > n/2 ? > > So you have a two node virtualization stuff, each hosting n/2 VMs, > and do the pacemaker clustering between those VMs? Yes. > > I'm sure you could easily add "somewhere else" a very bare bone VM > (or real) server, that is dedicated member of your cluster, but > never takes any resources? Just serves as arbitrator? as your "+1"? No, I'm afraid it is not that easy. There is simply is nothing that can be used. If there is anything, it is always available on both hosts/controllers. Imagine you would sell a standalone DRBD system (black box), that provides for example NFS to clients. You would want to have each and every additional service mirrored again. And you could not rely on additional customer NFS clients. > > May be easier, safer, and more transparent than > no-quorum=ignore plus some ping attribute based auto-shutdown. I agree on safer and transparent, but unfortunately, it not easier in our case. -- Bernd Schubert DataDirect Networks ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] pingd
On Fri, Sep 03, 2010 at 12:12:58PM +0200, Bernd Schubert wrote: > > > >> PS: (*) As you insist ;) on quorum with n/2 + 1 nodes, we use ping as > > > >> replacement. We simply cannot fulfill n/2 + 1, as controller failure > > > >> takes down 50% of the systems (virtual machines) and the systems > > > >> (VMs) of the 2nd controller are then supposed to take over failed > > > >> services. I see that n/2 + 1 is optimal and also required for a few > > > >> nodes. But if you have a larger set of system (e.g. minimum 6 with > > > >> the VM systems I have in my mind) n/2 + 1 is sufficient, IMHO. > > > > > > > > You meant to say you consider == n/2 sufficient, instead of > n/2 ? > > > > So you have a two node virtualization stuff, each hosting n/2 VMs, > > and do the pacemaker clustering between those VMs? > > Yes. > > > > > I'm sure you could easily add "somewhere else" a very bare bone VM > > (or real) server, that is dedicated member of your cluster, but > > never takes any resources? Just serves as arbitrator? as your "+1"? > > No, I'm afraid it is not that easy. There is simply is nothing that can be > used. If there is anything, it is always available on both hosts/controllers. > Imagine you would sell a standalone DRBD system (black box), that provides > for > example NFS to clients. You would want to have each and every additional > service mirrored again. And you could not rely on additional customer NFS > clients. > > > > > May be easier, safer, and more transparent than > > no-quorum=ignore plus some ping attribute based auto-shutdown. > > I agree on safer and transparent, but unfortunately, it not easier in our > case. Sell a Sheevaplug together with the two blackboxes, and tell them to plug it into somewhere. Yeah, well, maybe an other micro thingy with two ethernet adapters, so you can plug it into your fully redundant network topology ;-) Something like that. -- : Lars Ellenberg : LINBIT | Your Way to High Availability : DRBD/HA support and consulting http://www.linbit.com DRBD® and LINBIT® are registered trademarks of LINBIT, Austria. ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] pingd
On Fri, Sep 3, 2010 at 9:38 AM, Lars Ellenberg wrote: > On Thu, Sep 02, 2010 at 09:33:59PM +0200, Andrew Beekhof wrote: >> On Thu, Sep 2, 2010 at 4:05 PM, Lars Ellenberg >> wrote: >> > On Thu, Sep 02, 2010 at 11:00:12AM +0200, Bernd Schubert wrote: >> >> On Thursday, September 02, 2010, Andrew Beekhof wrote: >> >> > On Wed, Sep 1, 2010 at 11:59 AM, Bernd Schubert >> >> > > My proposal is to rip out all network code out of pingd and to add >> >> > > slightly modified files from 'iputils'. >> >> > >> >> > Close, but thats not portable. >> >> > Instead use ocf:pacemaker:ping which goes a step further and ditches >> >> > the daemon piece altogether. >> >> >> >> Hmm, we are already using that for now temporarily. But I don't think the >> >> ping >> >> RA is suitable for larger clusters. The ping script RA runs everything >> >> serially and only in intervals when called by lrmd. Now lets assume we >> >> have a >> >> 20 node cluster. >> >> >> >> nodes = 20 >> >> timeout = 2 >> >> attempts = 2 >> >> >> >> Makes 80s for a single run with default already rather small timeouts, >> >> which >> >> is IMHO a bit large. And with a shell script I don't see a way to improve >> >> that. While we could send the pings in parallel, I have no idea how to >> >> lock >> >> the variable of active nodes (active=`expr $active + 1`). I don't think >> >> that >> >> the simple sh or even bash have a semaphore or mutex lock. So IMHO, we >> >> need a >> >> language that supports that, rewriting the pingd RA is one choice, >> >> rewriting >> >> the ping RA into python is another. >> > >> > how about an fping RA ? >> > active=$(fping -a -i 5 -t 250 -B1 -r1 $host_list 2>/dev/null | wc -l) >> > >> > terminates in about 3 seconds for a hostlist of 100 (on the LAN, 29 of >> > which are alive). >> >> Happy to add if someone writes it :-) > > I thought so ;-) > Additional note to whomever is going to: > > With fping you can get fancy about "better connectivity", > you are not limited to the measure "number of nodes responding". > You could also use the statistics on packet loss and rtt provided on > stderr for -c or -C mode (example output below, chose what you think is > easier to parse), then do some scoring scheme on average or max packet loss, > rtt, or whatever else makes sense to you. > (If a switch starts dying, it may produce increasing packet loss first...) This sounds great. I think we want the ping RA to use fping where available. > > Or start a smokeping daemon, > and use the triggers there to change pacemaker attributes. > Uhm, well, thats probably no longer maintainable, though ;-) > > # fping -q -i 5 -t 250 -B1 -r2 -C5 -g 10.9.9.50 10.9.9.70 > 10.9.9.50 : 0.14 0.14 0.16 0.12 0.15 > 10.9.9.51 : - - - - - > 10.9.9.52 : - - - - - > 10.9.9.53 : 0.37 0.34 0.36 0.34 0.34 > 10.9.9.54 : 0.13 0.12 0.13 0.12 0.13 > 10.9.9.55 : 0.17 0.15 0.16 0.12 0.22 > 10.9.9.56 : 0.32 0.32 0.31 0.41 0.36 > 10.9.9.57 : 0.35 0.33 0.32 0.34 0.32 > 10.9.9.58 : - - - - - > 10.9.9.59 : - - - - - > 10.9.9.60 : - - - - - > 10.9.9.61 : - - - - - > 10.9.9.62 : - - - - - > 10.9.9.63 : - - - - - > 10.9.9.64 : - - - - - > 10.9.9.65 : 1.92 0.33 0.33 0.33 0.34 > 10.9.9.66 : - - - - - > 10.9.9.67 : - - - - - > 10.9.9.68 : - - - - - > 10.9.9.69 : 0.15 0.14 0.17 0.13 0.14 > 10.9.9.70 : - - - - - > > # fping -q -i 5 -t 250 -B1 -r2 -c5 -g 10.9.9.50 10.9.9.70 > 10.9.9.50 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.11/0.13/0.15 > 10.9.9.51 : xmt/rcv/%loss = 5/0/100% > 10.9.9.52 : xmt/rcv/%loss = 5/0/100% > 10.9.9.53 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.33/0.34/0.37 > 10.9.9.54 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.10/0.11/0.13 > 10.9.9.55 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.13/0.16/0.20 > 10.9.9.56 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.34/0.36/0.41 > 10.9.9.57 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.16/0.25/0.33 > 10.9.9.58 : xmt/rcv/%loss = 5/0/100% > 10.9.9.59 : xmt/rcv/%loss = 5/0/100% > 10.9.9.60 : xmt/rcv/%loss = 5/0/100% > 10.9.9.61 : xmt/rcv/%loss = 5/0/100% > 10.9.9.62 : xmt/rcv/%loss = 5/0/100% > 10.9.9.63 : xmt/rcv/%loss = 5/0/100% > 10.9.9.64 : xmt/rcv/%loss = 5/0/100% > 10.9.9.65 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.28/0.32/0.34 > 10.9.9.66 : xmt/rcv/%loss = 5/0/100% > 10.9.9.67 : xmt/rcv/%loss = 5/0/100% > 10.9.9.68 : xmt/rcv/%loss = 5/0/100% > 10.9.9.69 : xmt/rcv/%loss = 5/5/0%, min/avg/max = 0.13/0.14/0.15 > 10.9.9.70 : xmt/rcv/%loss = 5/0/100% > >> >> So in fact my first proposal also only was the first step - first add >> >> better >> >> network code and then to make it multi-threaded - each ping host gets its >> >> own >> >> thread. >> > >> > A working pingd daemon has the additional advantage that it can ask its >> > peers for their ping node count, before actually updating the attribute, >> > which should help with the "dampen race". >> >> That happens at the attrd level in both cases. pingd adds nothing here. > > I thought pingd did the dampening itself, even communicated with its peer > pingd's, and there was no more
Re: [Pacemaker] pingd
On Fri, Sep 3, 2010 at 12:12 PM, Bernd Schubert wrote: > On Friday, September 03, 2010, Lars Ellenberg wrote: >> > > how about an fping RA ? >> > > active=$(fping -a -i 5 -t 250 -B1 -r1 $host_list 2>/dev/null | wc -l) >> > > >> > > terminates in about 3 seconds for a hostlist of 100 (on the LAN, 29 of >> > > which are alive). >> > >> > Happy to add if someone writes it :-) >> >> I thought so ;-) >> Additional note to whomever is going to: >> >> With fping you can get fancy about "better connectivity", >> you are not limited to the measure "number of nodes responding". > > I think for the beginning, just the basic feature should be sufficient. > Actually I thought about to add an option to the existing ping RA to let the > user choose between ping and fping, it would default to ping. I will do that > mid of next week. Could you make fping the default binary if its installed and send us the patch when you\'re done? :-) > > >> You could also use the statistics on packet loss and rtt provided on >> stderr for -c or -C mode (example output below, chose what you think is >> easier to parse), then do some scoring scheme on average or max packet >> loss, rtt, or whatever else makes sense to you. >> (If a switch starts dying, it may produce increasing packet loss first...) > > That will require quite parsing, which I'm not comfortable with in a shell > script. I have no objections to later on add fping RA written in perl or > python. > > [...] > >> >> > >> PS: (*) As you insist ;) on quorum with n/2 + 1 nodes, we use ping as >> > >> replacement. We simply cannot fulfill n/2 + 1, as controller failure >> > >> takes down 50% of the systems (virtual machines) and the systems >> > >> (VMs) of the 2nd controller are then supposed to take over failed >> > >> services. I see that n/2 + 1 is optimal and also required for a few >> > >> nodes. But if you have a larger set of system (e.g. minimum 6 with >> > >> the VM systems I have in my mind) n/2 + 1 is sufficient, IMHO. >> > > >> > > You meant to say you consider == n/2 sufficient, instead of > n/2 ? >> >> So you have a two node virtualization stuff, each hosting n/2 VMs, >> and do the pacemaker clustering between those VMs? > > Yes. > >> >> I'm sure you could easily add "somewhere else" a very bare bone VM >> (or real) server, that is dedicated member of your cluster, but >> never takes any resources? Just serves as arbitrator? as your "+1"? > > No, I'm afraid it is not that easy. There is simply is nothing that can be > used. If there is anything, it is always available on both hosts/controllers. > Imagine you would sell a standalone DRBD system (black box), that provides for > example NFS to clients. You would want to have each and every additional > service mirrored again. And you could not rely on additional customer NFS > clients. > >> >> May be easier, safer, and more transparent than >> no-quorum=ignore plus some ping attribute based auto-shutdown. > > I agree on safer and transparent, but unfortunately, it not easier in our > case. > > -- > Bernd Schubert > DataDirect Networks > > ___ > Pacemaker mailing list: Pacemaker@oss.clusterlabs.org > http://oss.clusterlabs.org/mailman/listinfo/pacemaker > > Project Home: http://www.clusterlabs.org > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf > Bugs: > http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker > ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] pingd
On Fri, Sep 03, 2010 at 12:12:58PM +0200, Bernd Schubert wrote: > On Friday, September 03, 2010, Lars Ellenberg wrote: > > > > how about an fping RA ? > > > > active=$(fping -a -i 5 -t 250 -B1 -r1 $host_list 2>/dev/null | wc -l) > > > > > > > > terminates in about 3 seconds for a hostlist of 100 (on the LAN, 29 of > > > > which are alive). > > > > > > Happy to add if someone writes it :-) > > > > I thought so ;-) > > Additional note to whomever is going to: > > > > With fping you can get fancy about "better connectivity", > > you are not limited to the measure "number of nodes responding". > > I think for the beginning, just the basic feature should be sufficient. > Actually I thought about to add an option to the existing ping RA to let the > user choose between ping and fping, it would default to ping. I will do that > mid of next week. > > > > You could also use the statistics on packet loss and rtt provided on > > stderr for -c or -C mode (example output below, chose what you think is > > easier to parse), then do some scoring scheme on average or max packet > > loss, rtt, or whatever else makes sense to you. > > (If a switch starts dying, it may produce increasing packet loss first...) > > That will require quite parsing, which I'm not comfortable with in a shell > script. I have no objections to later on add fping RA written in perl or > python. -s causes a summary to be displayed, like so: fping -s -q -c 2 $host_list 21 targets 6 alive 15 unreachable 0 unknown addresses 0 timeouts (waiting for response) 42 ICMP Echos sent 12 ICMP Echo Replies received 0 other ICMP received 0.11 ms (min round trip time) 0.47 ms (avg round trip time) 2.22 ms (max round trip time) 2.219 sec (elapsed real time) That is easy enough to parse even from a shell script. You just chose if you want to use "numbers alive", or "number replies received", the latter accounting for packet loss. If you want, you can resort to awk ;-) N_REPLIES=$(fping -s -q -c $N_PINGS $host_list 2>&1 | awk '/ICMP Echo Replies received/ { print $1 }') # update the attribute with $N_REPLIES, instead of number alive, # maybe scale with N_PINGS and/or N_HOSTS ... -- : Lars Ellenberg : LINBIT | Your Way to High Availability : DRBD/HA support and consulting http://www.linbit.com ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] Setting up routing for a virtual ip
Original-Nachricht > Datum: Thu, 02 Sep 2010 19:08:00 +0200 > Von: "Stephan-Frank Henry" > An: The Pacemaker cluster resource manager > Betreff: Re: [Pacemaker] Setting up routing for a virtual ip > Original-Nachricht > > Datum: Thu, 02 Sep 2010 11:40:13 +0200 > > Von: "Stephan-Frank Henry" > > An: pacemaker@oss.clusterlabs.org > > Betreff: [Pacemaker] Setting up routing for a virtual ip > > > Hello everyone, > > > > I am currently stuck trying to set up routing for a configured virtual > ip > > to the static ip on the same host. > > > > static ip: 150.158.1.2 > > (I have two nics in use, but this is the important one) > > virtual ip: 1.2.3.4 > > nic: eth0 > > > > versions: > > Debian Lenny 2.6.33.3 x86_64 > > corosync : 1.2.1-1 > > libheartbeat2 : 3.0.3-2 > > > > Here are the relevant parts: > > > provider="heartbeat"> > > > > > > > > > >> value="22"/> > > > > > > > > > > > > > > provider="heartbeat"> > > > > > > > >value="150.158.1.2"/> > >> value="0.0.0.0/0"/> > > > > > > > > > > > name="monitor"/> > > > > > > > > But when I run it, it prints out messages like: > > Route[25503]: ERROR: ip_gateway Failed to add network route: to > 0.0.0.0/0 > > via 150.158.1.2 src 1.2.3.5 > > WARN: unpack_rsc_op: Processing failed op ip_gateway_start_0 on > nodealpha: > > unknown error (1) > > > > I have tried it with variations (f.i. leaving out the device) but > without > > success. > > > > If I remove the routing config, it works fine. > > > > What am I missing? > > > > Could it be related to the fact that I do not see a virtual interface > via > > ifconfig (-a)? > > > > thanks > > > > Frank > > Self-update :D > > I have updated the settings and now am only using IPaddr instead of the *2 > version and now at least I can see the virtual ip. > > I also commented out the device (dunno if I should put in eth0 or eth0:0) > and changed the virtual ip to 150.158.1.5 > > now I am getting this error > > crmd: [20637]: info: do_lrm_rsc_op: Performing > key=35:1:0:d010917f-1f67-415a-b02b-97c784c1974f op=ip_gateway_start_0 ) > lrmd: [20634]: info: rsc:ip_gateway:15: start > crmd: [20637]: info: te_rsc_command: Initiating action 35: start > ip_gateway_start_0 on nodealpha (local) > crmd: [20637]: info: process_lrm_event: LRM operation > ip_resource_monitor_1 (call=14, rc=0, cib-update=41, confirmed=false) ok > lrmd: [20634]: info: RA output: (ip_gateway:start:stderr) RTNETLINK > answers: File exists > crmd: [20637]: info: match_graph_event: Action ip_resource_monitor_1 > (34) confirmed on nodealpha (rc=0) > Route[21137]: ERROR: ip_gateway Failed to add network route: to 0.0.0.0/0 > via 150.158.1.2 > crmd: [20637]: info: process_lrm_event: LRM operation ip_gateway_start_0 > (call=15, rc=1, cib-update=42, confirmed=true) unknown error > > Anyone? > > BTW: I'm not a network expert, so please highlight any issues. > > thanks > > Frank Oookay, so it looks like one of the issues is my near complete lack of knowledge of networking. From one of our resident networking guys it seems as though I need iptables to solve this issue. Something like iptables -t nat -A PREROUTING -p tcp -i eth0 -d 150.158.1.5 -j DNAT --to 150.158.1.2 iptables -A FORWARD -p tcp -i eth0 -d 150.158.1.5 -j ACCEPT Is there any way to natively embed this into the resource management or do I need to have my own scripts? I checked the files but I could not really find anything useful. thanks frank -- GMX DSL SOMMER-SPECIAL: Surf & Phone Flat 16.000 für nur 19,99 Euro/mtl.!* http://portal.gmx.net/de/go/dsl ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] Node doesn't rejoin automatically after reboot
Tom Tux wrote: If I disjoin one clusternode (node01) for maintenance-purposes (/etc/init.d/openais stop) and reboot this node, then it will not join himself automatically into the cluster. After the reboot, I have the following error- and warn-messages in the log: Sep 3 07:34:15 node01 mgmtd: [9202]: info: login to cib failed: live Do you have messages like this, too? Aug 30 15:48:10 xen-test1 corosync[5851]: [IPC ] Invalid IPC credentials. Aug 30 15:48:10 xen-test1 cib: [5858]: info: init_ais_connection: Connection to our AIS plugin (9) failed: unknown (100) Aug 30 15:48:10 xen-test1 cib: [5858]: CRIT: cib_init: Cannot sign in to the cluster... terminating http://news.gmane.org/find-root.php?message_id=%3c4C7C0EC7.2050708%40cbnco.com%3e Mike ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
[Pacemaker] MCP init script to 21/79?
On 08/24/2010 11:06 PM, Andrew Beekhof wrote: On Wed, Aug 25, 2010 at 8:02 AM, Vladislav Bogdanov wrote: 25.08.2010 08:56, Andrew Beekhof wrote: On Wed, Aug 25, 2010 at 7:39 AM, Vladislav Bogdanov wrote: Hi all, pacemaker has # chkconfig - 90 90 in its MCP initscript. Shouldn't it be corrected to 90 10? I thought higher numbers started later and shut down earlier... no? Nope, they are in a natural order for both start and stop sequences. So lower number means 'do start or stop earlier'. grep '# chkconfig' /etc/init.d/* Ok, thanks. Changed to 10 Given that corosync default is 20/80, shouldnt mcp be 21/79? Regards -steve ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] MCP init script to 21/79?
03.09.2010 19:34, Steven Dake wrote: >>> Nope, they are in a natural order for both start and stop sequences. >>> So lower number means 'do start or stop earlier'. >>> >>> grep '# chkconfig' /etc/init.d/* >>> >> >> Ok, thanks. Changed to 10 >> > > Given that corosync default is 20/80, shouldnt mcp be 21/79? I think that pcmk may require additional services to be started (I at least see reference to cooperation with cman for GFS as one of pcmk MCP scenarios in Andrew's wiki, but that scenario is still unclear to me), so it is safer to have it start later, 90 is ok for me. That is also what Vadim wrote about. Best, Vladislav ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] MCP init script to 21/79?
On 09/03/2010 09:56 AM, Vladislav Bogdanov wrote: 03.09.2010 19:34, Steven Dake wrote: Nope, they are in a natural order for both start and stop sequences. So lower number means 'do start or stop earlier'. grep '# chkconfig' /etc/init.d/* Ok, thanks. Changed to 10 Given that corosync default is 20/80, shouldnt mcp be 21/79? I think that pcmk may require additional services to be started (I at least see reference to cooperation with cman for GFS as one of pcmk MCP scenarios in Andrew's wiki, but that scenario is still unclear to me), so it is safer to have it start later, 90 is ok for me. That is also what Vadim wrote about. Best, Vladislav I was mistaken, not having read the current code. Ignore the noise. Regards -steve ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
[Pacemaker] Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to exist
I was following cluster from scratch guide and everything were fine until i get here: http://www.clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Clusters_from_Scratch/ch08s05.html The command: grep ERROR: /var/log/messages | grep -v unpack_resources Says: Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to exist Crm_mon: WebFS:0_start_0 (node=Master, call=53, rc=5, status=complete): not installed WebFS:1_start_0 (node=HPHDX, call=46, rc=5, status=complete): not installed I double checked and certified that everything the guide said to install i did: yum install -y drbd-pacemaker yum install -y gfs2-utils gfs-pcmk Is something i didn't install? the directory -> /dev/drbd/by-res/wwwdata does not exist, i think it should... Any hint? ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to exist
On 09/03/2010 01:05 PM, Alisson Landim wrote: > I was following cluster from scratch guide and everything were fine > until i get here: > > http://www.clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Clusters_from_Scratch/ch08s05.html > > The command: > grep ERROR: /var/log/messages | grep -v unpack_resources > > Says: > Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to > existdation.org/enter_bug.cgi?product=Pacemaker Don't know your DRBD version, but make sure you've got the corresponding udev stuff installed. If not, replace references to /dev/drbd/by-res/* with their actual device nodes (/dev/drbd0, for example) -- : Adam Gandelman : LINBIT | Your Way to High Availability : Telephone: 503-573-1262 ext. 203 : Sales: 1-877-4-LINBIT / 1-877-454-6248 : : 7959 SW Cirrus Dr. : Beaverton, OR 97008 : : http://www.linbit.com ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
Re: [Pacemaker] Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to exist
Adam Gandelman wrote: "Don't know your DRBD version," yum install drbd-pacemaker returns: Package drbd-pacemaker-8.3.7-2.fc13.x86_64 already installed and latest version also drbdadm says: Version: 8.3.7 (api:88) Adam Gandelman wrote: "but make sure you've got the corresponding udev stuff installed" How do i check that? Adam Gandelman wrote:If not, replace references to /dev/drbd/by-res/* with their actual device nodes (/dev/drbd0, for example) Ok, i replaced with drbd1 which is my device and got on var/log/messages: crmd: [28175]: ERROR: process_lrm_event: Op dlm:1_start_0 (call=10): Cancelled crmd: [28175]: ERROR: do_lrm_invoke: Not creating resource for a delete event: (null) Filesystem[31537]: ERROR: Couldn't mount filesystem /dev/drbd1 on /var/www/html crmd: [30593]: ERROR: process_lrm_event: LRM operation WebFS:1_stop_0 (24) Timed Out (timeout=10ms) And my system broadcast a message of kernel bug: Package:kernel Latest Crash:Fri 03 Sep 2010 07:44:56 PM Command:not_applicable Reason: kernel BUG at fs/dlm/lock.c:242! Comment:None Bug Reports: kernel:invalid opcode: [#1] SMP kernel:last sysfs file: /sys/kernel/dlm/web/event_done kernel:Stack: kernel:Call Trace: kernel:Code: 8b e0 00 00 00 8b 73 3c 44 8b 83 d0 00 00 00 48 c7 c7 38 c7 18 a0 31 c0 e8 85 f7 2c e1 48 c7 c7 6d c7 18 a0 31 c0 e8 77 f7 2c e1 <0f> 0b eb fe 59 0f 95 c0 0f b6 c0 5b c9 c3 55 48 89 e5 53 48 83 Date: Fri, 3 Sep 2010 15:16:13 -0700 From: adam.gandel...@linbit.com To: pacemaker@oss.clusterlabs.org Subject: Re: [Pacemaker] Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to exist Message body On 09/03/2010 01:05 PM, Alisson Landim wrote: I was following cluster from scratch guide and everything were fine until i get here: http://www.clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Clusters_from_Scratch/ch08s05.html The command: grep ERROR: /var/log/messages | grep -v unpack_resources Says: Couldn't find device [/dev/drbd/by-res/wwwdata]. Expected /dev/??? to existdation.org/enter_bug.cgi?product=Pacemaker Don't know your DRBD version, but make sure you've got the corresponding udev stuff installed. If not, replace references to /dev/drbd/by-res/* with their actual device nodes (/dev/drbd0, for example) -- : Adam Gandelman : LINBIT | Your Way to High Availability : Telephone: 503-573-1262 ext. 203 : Sales: 1-877-4-LINBIT / 1-877-454-6248 : : 7959 SW Cirrus Dr. : Beaverton, OR 97008 : : http://www.linbit.com ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker ___ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker