Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
sanboot:ubuntu_iscsi [2010/08/11 12:21]
cinquero
sanboot:ubuntu_iscsi [2010/08/16 20:55] (current)
cinquero
Line 213: Line 213:
 find /etc -name '​S*open-iscsi'​ find /etc -name '​S*open-iscsi'​
 </​code>​ </​code>​
 +  * One possibly has to repeat the previous step when upgrading/​reinstalling open-iscsi.
   * create /​etc/​init/​open-iscsi.conf:​   * create /​etc/​init/​open-iscsi.conf:​
 <​code>​ <​code>​
Line 218: Line 219:
  
 start on filesystem start on filesystem
 +# no stopping!
 exec /​etc/​init.d/​open-iscsi start exec /​etc/​init.d/​open-iscsi start
 </​code>​ </​code>​
Line 226: Line 228:
 </​code>​ </​code>​
  
-=== Stabilization/​Paranoia ===+=== Stabilization/​Paranoia/​Warnings ​===
  
 in order to prevent data loss (for example, a corrupted dpkg db), one may do the following stuff: in order to prevent data loss (for example, a corrupted dpkg db), one may do the following stuff:
  
   * Enable data journaling. That will cut your maximum write performance in half.   * Enable data journaling. That will cut your maximum write performance in half.
-  * Increase ​node.session.timeo.replacement_timeout in iscsid.conf to avoid escalation of temporary connection problems to upper layers. In *my* situation it *never* makes sense to tell the ext4 driver about I/O problems because there is no fallback. It would just give me a corrupted ​filesystem.+  * One may increase ​node.session.timeo.replacement_timeout in iscsid.conf to avoid escalation of temporary connection problems to upper layers. In *my* situation it (almost) ​*never* makes sense to tell the ext4 driver about I/O problems because there is no fallback. It would just give me an unclean filesystem (not unmounted properly, like hard reset). However(!!),​ an unclean ​filesystem ​may be better than a totally screwed one: imagine your iSCSI target host crashes, looses some data that still has not been written to disk, and the client continues to use that target without doing a journal replay first... personally, I try to avoid that situation by never starting the iSCSI target automatically (removed from init scripts, starting it manually after killing the client machines...). IMHO the iSCSI protocol is a bit dumb. The target should detect a crash and refuse session continuation after restart... again, I'm not liable for *any* sort of data loss if you act according to these explanations. You *will* screw your data. Sooner or later. 
 +  * Additional warning: SAN boot using gpxe seems to not allow using a MaxSessions=1/​MaxConnections=1 setting for the target, which would prevent concurrent accesses -- a very basic need for data integrity purposes.

QR Code
QR Code sanboot:ubuntu_iscsi (generated for current page)