Hello all, I am again asking this question. If someone helps, it would be great !
We are planning to have a 2 node HA solution which is also a DR mechanism. So the two nodes will be different location - can be on a WAN. The plan is to have Active-Active node which runs the services independantly, but if one of the node A fails or the network(WAN) fails, the other node B should take over the service of A. We can have a intelligent client which will shift the service to B if it cannot communicate to A. Since the two nodes will be on a WAN, chances of split brain remain. So what can we do to reduce this 1) Should we have a quorum server ? Should it be in different location ? Is there any configuration/sample/tutorial/pointers if someone has used this ? 2) if not a quorum server, what are the alternatives ? As I have wrote before, we have the non-WAN 2 node HA working to a extent.(still a prototype). Pls Help. Let me know if more info is needed. Thanks. -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Nagaraja, Madhukar (Madhukar) Sent: Tuesday, February 24, 2009 2:47 PM To: [email protected] Subject: [Linux-HA] Re: Split Site. To come out of the split-brain, is quorum the choice ? Also if we have to use DRBD on this 2 node split-site, will there be issues. Basically we will be using 1) our product 2) Postgresql/DRBD or datareplication Initially in the demo we had a IP and MAC address takeover also, but in the split-site scenario we wont be having it. Any pointers appreciated. Thanks. -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of [email protected] Sent: Saturday, February 21, 2009 12:30 AM To: [email protected] Subject: Linux-HA Digest, Vol 63, Issue 45 Send Linux-HA mailing list submissions to [email protected] To subscribe or unsubscribe via the World Wide Web, visit http://lists.linux-ha.org/mailman/listinfo/linux-ha or, via email, send a message with subject or body 'help' to [email protected] You can reach the person managing the list at [email protected] When replying, please edit your Subject line so it is more specific than "Re: Contents of Linux-HA digest..." Today's Topics: 1. Split Site. (Nagaraja, Madhukar (Madhukar)) 2. Ted Kotlowski is out of the office. ([email protected]) 3. Re: Split Site. (Alex Balashov) 4. Re: Split Site. (Alex Balashov) 5. Re: Secondary server Single point of failure (Dejan Muhamedagic) ---------------------------------------------------------------------- Message: 1 Date: Fri, 20 Feb 2009 14:53:59 +0530 From: "Nagaraja, Madhukar (Madhukar)" <[email protected]> Subject: [Linux-HA] Split Site. To: General Linux-HA mailing list <[email protected]> Message-ID: <e9f099885b445541a55d830fc62e175a06627a2...@inbansxchmbsa3.in.alcatel-lucent.com> Content-Type: text/plain; charset="us-ascii" Hello, There is a requirment for us to have 2 node HA, but the thing is each of the node will be in a different location. I guess they can communicate thru WAN and the IP address will be pingable to the other. One will be active and the other passive. Only when one goes down the other node in a diff location will take over. I had prototyped a 2 node HA with both the nodes on the same LAN. But now they want it in different locations. Can I use the same thing for this or do I need to use quorum or some other mechanism ?? If quorum, can you please point out to some info who have done this ? I searched the net but could not get anything which could answer me. Please let me know. Thanks. ------------------------------ Message: 2 Date: Fri, 20 Feb 2009 05:02:03 -0500 From: [email protected] Subject: [Linux-HA] Ted Kotlowski is out of the office. To: General Linux-HA mailing list <[email protected]> Message-ID: <of2ca86d53.00081792-on85257563.00371eca-85257563.00371...@assurant.com> Content-Type: text/plain; charset=US-ASCII I will be out of the office starting 02/20/2009 and will not return until 02/23/2009. I will respond to your message when I return. If your request requires immediate attention, Please contact the MVS Technical Support Hotline at 1-866-866-4488 x12000 ************************************************************************************** This e-mail message and all attachments transmitted with it may contain legally privileged and/or confidential information intended solely for the use of the addressee(s). If the reader of this message is not the intended recipient, you are hereby notified that any reading, dissemination, distribution, copying, forwarding or other use of this message or its attachments is strictly prohibited. If you have received this message in error, please notify the sender immediately and delete this message and all copies and backups thereof. Thank you. ************************************************************************************** ------------------------------ Message: 3 Date: Fri, 20 Feb 2009 05:06:28 -0500 From: Alex Balashov <[email protected]> Subject: Re: [Linux-HA] Split Site. To: General Linux-HA mailing list <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1; format=flowed You want to have two nodes separated by Internet topology? Sure, there's no problem with that. Why wouldn't you be able to use the same setup? The problem is, it's very, _very_ easy to end up with a split brain condition. If anything at all out of the ordinary happens to the connectivity between the two sites - e.g. route flaps, congestion, unusually high latency, outages - it's very easy for the two nodes to think the other died. Nagaraja, Madhukar (Madhukar) wrote: > Hello, > > There is a requirment for us to have 2 node HA, but the thing is each of the > node will be in a different location. I guess they can communicate thru WAN > and the IP address will be pingable to the other. One will be active and the > other passive. Only when one goes down the other node in a diff location will > take over. I had prototyped a 2 node HA with both the nodes on the same LAN. > But now they want it in different locations. Can I use the same thing for > this or do I need to use quorum or some other mechanism ?? > > If quorum, can you please point out to some info who have done this ? I > searched the net but could not get anything which could answer me. > > Please let me know. > > Thanks. > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems -- Alex Balashov Evariste Systems Web : http://www.evaristesys.com/ Tel : (+1) (678) 954-0670 Direct : (+1) (678) 954-0671 Mobile : (+1) (678) 237-1775 ------------------------------ Message: 4 Date: Fri, 20 Feb 2009 05:07:00 -0500 From: Alex Balashov <[email protected]> Subject: Re: [Linux-HA] Split Site. To: General Linux-HA mailing list <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Of course, you won't be able to share an IP address among the nodes. Well, unless you've got some sort of clever VPN setup to pull that off. Alex Balashov wrote: > You want to have two nodes separated by Internet topology? Sure, > there's no problem with that. Why wouldn't you be able to use the > same setup? > > The problem is, it's very, _very_ easy to end up with a split brain > condition. If anything at all out of the ordinary happens to the > connectivity between the two sites - e.g. route flaps, congestion, > unusually high latency, outages - it's very easy for the two nodes to > think the other died. > > Nagaraja, Madhukar (Madhukar) wrote: > >> Hello, >> >> There is a requirment for us to have 2 node HA, but the thing is each >> of the node will be in a different location. I guess they can >> communicate thru WAN and the IP address will be pingable to the other. >> One will be active and the other passive. Only when one goes down the >> other node in a diff location will take over. I had prototyped a 2 >> node HA with both the nodes on the same LAN. But now they want it in >> different locations. Can I use the same thing for this or do I need >> to use quorum or some other mechanism ?? >> >> If quorum, can you please point out to some info who have done this ? >> I searched the net but could not get anything which could answer me. >> >> Please let me know. >> >> Thanks. >> _______________________________________________ >> Linux-HA mailing list >> [email protected] >> http://lists.linux-ha.org/mailman/listinfo/linux-ha >> See also: http://linux-ha.org/ReportingProblems > > -- Alex Balashov Evariste Systems Web : http://www.evaristesys.com/ Tel : (+1) (678) 954-0670 Direct : (+1) (678) 954-0671 Mobile : (+1) (678) 237-1775 ------------------------------ Message: 5 Date: Fri, 20 Feb 2009 12:29:01 +0100 From: Dejan Muhamedagic <[email protected]> Subject: Re: [Linux-HA] Secondary server Single point of failure To: [email protected], General Linux-HA mailing list <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=us-ascii Hi, On Fri, Feb 13, 2009 at 12:21:34PM +0530, Manjunath K.P wrote: > > Hi, > > My setup involves a shared storage connected to two Servers running heartbeat > V 2. Server A is Primary and Server B is secondary. The failover happens fine > when I bring down the Primary server, The resources ( Shared Disk and > services ) switch over to Secondary server Fine. And When the Primary server > comes back up, It takes over the control back. > > But Weird problem I am facing is that, When the secondary ( Server B ) Goes > down, Primary server (Server A) looses all its resources ( Shared Disk and > service stops ), Till the Server B comes back up. > > So technically My Secondary server has become a single point of failure, As > long as its up everything works once It goes down everything is down. > > Can anyone explain me what I am doing wrong here. Perhaps if you show the logs. > Here is my haresource file. > > Server A:- serverA Filesystem::/dev/sda1::/vz::ext3 vzup > > Server B:- serverA Filesystem::/dev/sda1::/vz1::ext3 vzup > > Please let me know If I need to furnish any further details. > > P.S:- haresource file has some difference in both the servers ( Mount > point vz in Server A and mount point vz1 in Server B. > Thats intentional ). haresources should be the same on both nodes. Thanks, Dejan > Regards, > Manjunath > > > Add more friends to your messenger and enjoy! Go to > http://messenger.yahoo.com/invite/ > > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems ------------------------------ _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems End of Linux-HA Digest, Vol 63, Issue 45 **************************************** _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
