[PATCH] Allow update of discovery records.

2009-09-30 Thread Hannes Reinecke
We should allow updating of discovery records, too. This enables open-iscsi to use different settings during discovery to different targets. Signed-off-by: Hannes Reinecke diff --git a/usr/idbm.c b/usr/idbm.c index ee38e64..ca5479e 100644 --- a/usr/idbm.c +++ b/usr/idbm.c @@ -2097,6 +2097,35 @

MSA2012i CHAP parameter

2009-09-30 Thread Hannes Reinecke
Hi all, just to share with you some more tidbits: If you enable 'CHAP' with a MSA2012i, the username is actually _not_ empty (like eg the MPX100 required it), but rather the initiatorname. And the Name and Secret for mutual CHAP are username_in and password_in. I probably should put this out i

Re: SD Driver incrementing on every snapshot.

2009-09-30 Thread Mike Christie
On 09/29/2009 06:58 PM, Matthew Schumacher wrote: > Group, > > I'm using linux to backup an ISCSI target. My backup script calls the > snapshot function on my iscsi server, then uses open-iscsi to connect to > a snapshot target on that server. The target is backed up, then I > logoff the target

Re: [PATCH] Allow empty username for CHAP

2009-09-30 Thread Mike Christie
On 09/30/2009 01:32 AM, Hannes Reinecke wrote: > Mike Christie wrote: >> On 09/29/2009 08:46 AM, Hannes Reinecke wrote: >>> Some iSCSI implementations (eg HP) is using an empty username for >>> CHAP negotiations. So we should be allowing the same. >>> >> >> Do we need this support for discovery? T

Re: MSA2012i CHAP parameter

2009-09-30 Thread Mike Christie
On 09/30/2009 06:11 AM, Hannes Reinecke wrote: > Hi all, > > just to share with you some more tidbits: > > If you enable 'CHAP' with a MSA2012i, the username > is actually _not_ empty (like eg the MPX100 required it), > but rather the initiatorname. Do you think we should put target info in some

Re: [PATCH] Allow update of discovery records.

2009-09-30 Thread Mike Christie
On 09/30/2009 03:56 AM, Hannes Reinecke wrote: > We should allow updating of discovery records, too. > This enables open-iscsi to use different settings during > discovery to different targets. > > Signed-off-by: Hannes Reinecke > Thanks. Merged in 3bb88ea9be814a2f598cf6f1d1fcdbc2f57ff0a8. See m

Suse 10.2 install of open-iscsi 2.0-871

2009-09-30 Thread Taylor
I am getting compile errros when running make KSRC=/usr/src/linux KBUILD_OUTPUT=/usr/src/linux-obj/x86_64/smp on a Suse 10.2 machine. I made sure the Cryptographic API and CRC32 options were set to Yes in kernel. I believe the problem is the 2.6.16-suse.patch isn't being run before the compile

[bug-report] no error reporting when network interface was changed

2009-09-30 Thread eliasp
Hi, during a reconfiguration of our servers, I removed a networkinterface (br2). This networkinterface was previously used by open-iscsi (iface.net_ifacename = br2). I didn't remember having set this interface in the open-iscsi configuration and so I ran into an error not being able to reconnect

Re: Suse 10.2 install of open-iscsi 2.0-871

2009-09-30 Thread Mike Christie
On 09/30/2009 11:07 AM, Taylor wrote: > I am getting compile errros when running make KSRC=/usr/src/linux > KBUILD_OUTPUT=/usr/src/linux-obj/x86_64/smp on a Suse 10.2 machine. > > I made sure the Cryptographic API and CRC32 options were set to Yes in > kernel. > > I believe the problem is the 2.6

RE: SD Driver incrementing on every snapshot.

2009-09-30 Thread Lewis, Karl
I'm experimenting with something similar, but it's still in the early planning. In my case, I have several hosts logging into a common target, writing file backups to a directory, then logging out. I planned to snap the backup LUN on the iSCSI target, then mount the snapshot LUN for backup pu