Administrace komentářů

mam nastavit HA cluster ze dvou serveru (box423 a box402) - bohuzel se v tom nevyznam. Sel jsem podle nejakych tutorialu a ve vysledku to _vetsinou_ chodi - apache a mysql bezi na jednom, nebo druhem, ale rozhodne jsou dostupne na spolecne adrese.

Problem je, ze kdyz stridave rebootuju (prikazem reboot - cili velmi nezne) ty dva servery, tak se mi kazdou chvili objevi split-brain, nebo se jeden zacne tvarit jako diskless a ja netusim proc. Trikrat to projde, po ctvrte nikoli (aniz bych videel nejakou zavislost).

IMHO mam spatne konfiguraci, ale nevim v cem. Zamerem je, ze na jednom serveru bezi vse a druhy se flaka, pokud prvni zdechne zcela ci castecne, tak druhy prebere vse. V praxi by to nastat nemelo, krome pripadu testovani prave teto vlastnosti (HW spolehlivy, zatez smesna, pozadavky minimalni, jinde nam to bezi na single stroji s uptimem nekolik let, ale zakaznik proste touzi po redundanci, kvuli pocitu sucha a bezpeci - dokonce ani o data moc nejde, z principu veci jsou zalohpvana na spouste mist)

box423 ~ # cat /proc/drbd
version: 8.3.6 (api:88/proto:86-91)
GIT-hash: f3606c47cc6fcf6b3f086e425cb34af8b7a81bbf build by @box423, 2011-12-22 11:43:07

 1: cs:Connected ro:Primary/Secondary ds:UpToDate/Diskless C r----
    ns:3857 nr:296 dw:7684 dr:25109 al:8 bm:574 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:521208

box423 ~ # crm status
============
Last updated: Fri Dec 30 20:32:06 2011
Last change: Fri Dec 30 20:22:16 2011
Stack: openais
Current DC: box423 - partition with quorum
Version: 1.1.6-0c7312c689715e096b716419e2ebc12b57962052
2 Nodes configured, 2 expected votes
6 Resources configured.
============

Online: [ box423 box402 ]

 Master/Slave Set: WebDataClone [WebData]
     Masters: [ box423 ]
     Slaves: [ box402 ]
 Resource Group: mysql
     WebFS	(ocf::heartbeat:Filesystem):	Started box423
     ClusterIP	(ocf::heartbeat:IPaddr2):	Started box423
     mysqld	(lsb:mysql):	Started box423
     WebSite	(lsb:apache2):	Started box423

box423 ~ # crm configure show
node box402 \
	attributes standby="off"
node box423 \
	attributes standby="off"
primitive ClusterIP ocf:heartbeat:IPaddr2 \
	params ip="10.0.1.222" cidr_netmask="32" \
	op monitor interval="30s"
primitive WebData ocf:linbit:drbd \
	params drbd_resource="shared_data" \
	op monitor interval="60s"
primitive WebFS ocf:heartbeat:Filesystem \
	params device="/dev/drbd/by-res/shared_data" directory="/home" fstype="ext3"
primitive WebSite lsb:apache2 \
	op monitor interval="1min"
primitive mysqld lsb:mysql
group mysql WebFS ClusterIP mysqld WebSite
ms WebDataClone WebData \
	meta master-max="1" master-node-max="1" clone-max="2" clone-node-max="1" notify="true" target-role="Started"
location prefer-box423 WebSite 50: box423
colocation WebSite-with-WebFS inf: WebSite WebFS
colocation fs_on_drbd inf: WebFS WebDataClone:Master
colocation website-with-ip inf: WebSite ClusterIP
order WebFS-after-WebData inf: WebDataClone:promote WebFS:start
order WebSite-after-WebFS inf: WebFS WebSite
order apache-after-ip inf: ClusterIP WebSite
property $id="cib-bootstrap-options" \
	dc-version="1.1.6-0c7312c689715e096b716419e2ebc12b57962052" \
	cluster-infrastructure="openais" \
	expected-quorum-votes="2" \
	stonith-enabled="false" \
	no-quorum-policy="ignore"
rsc_defaults $id="rsc-options" \
	resource-stickiness="100"

box423 ~ # cat /etc/drbd.d/shared_data.res 
resource shared_data {
  meta-disk internal;
  device    /dev/drbd1;
  syncer {
    verify-alg sha1;
  }
  net {
    allow-two-primaries;
  }
  on box423 {
    device    /dev/drbd1;
    disk      /dev/sda4;
    address   10.0.1.104:7789;
  }
  on box402 {
    device    /dev/drbd1;
    disk      /dev/sda4;
    address   10.0.1.103:7789;
  }
}