Re: Keepalived backup failed to become master
Paul Robert Marino <[email protected]>
| Newsgroups | gmane.linux.keepalived.devel |
|---|---|
| Message-ID | <CAPJdpdDNBjxHeVp-Arhb4HJ8f-1G22Rn2g1OGPpS4gM4O_uRwQ@mail.gmail.com> |
correction set them both to "BACKUP" On Sun, Feb 2, 2014 at 12:12 PM, Paul Robert Marino <[email protected]> wrote: > Set both to SLAVE. > They will then have an election for the master based on the priorities > > The docs that say to set one to master and the other to slave are old and > config produced based on them can exhibit erratic behaviours during > failover. > > -- Sent from my HP Pre3 > > ________________________________ > On Jan 31, 2014 13:19, Tiago Cruz <[email protected]> wrote: > > Not yet! > > Sound like my master become crazy. The machine is (kind of) alive: > > # tcpdump -ennqti any arp |egrep '200|xd\:6x' > tcpdump: verbose output suppressed, use -v or -vv for full protocol decode > listening on any, link-type LINUX_SLL (Linux cooked), capture size 65535 > bytes > B 00:xe:4f:xx:xd:6x Request who-has 200.xxx.xxx.193 tell 200.xxx.xxx.95, > length 46 > B 00:xe:4f:xx:xd:6x Request who-has 200.xxx.xxx.160 tell 200.xxx.xxx.95, > length 46 > B 00:xe:4f:xx:xd:6x Request who-has 200.xxx.xxx.100 tell 200.xxx.xxx.95, > length 46 > > This machine 200.xxx.xxx.95 supposed to be dead! I can't ping or ssh into, > but she still as a zombie around :) > > > On Fri, Jan 31, 2014 at 3:34 PM, Ryan O'Hara <[email protected]> wrote: >> >> No problem. Did it fix the problem you were having? > > -- > -- Tiago Cruz > ------------------------------------------------------------------------------ Managing the Performance of Cloud-Based Applications Take advantage of what the Cloud has to offer - Avoid Common Pitfalls. Read the Whitepaper. http://pubads.g.doubleclick.net/gampad/clk?id=121051231&iu=/4140/ostg.clktrk