Re: Best way to create multiple TCP flows on 10 Gbps link

2014-08-27 Thread Mark Lehrer

On Tue, 26 Aug 2014 13:05:11 -0700
 Learner learner.st...@gmail.com wrote:

How many iscsi and underlying top sessions are u using? If multiple,
pls check if all to sessions are being used.

Btw, what tuning did u perform to fix Tcp BDP issue?


I'm just doing netcat tests to/from /dev/shm at the moment.

I wouldn't consider it fixed necessarily, but the info from this link was 
useful:


http://www.kehlet.cx/articles/99.html

Thanks,
Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to open-iscsi+unsubscr...@googlegroups.com.
To post to this group, send email to open-iscsi@googlegroups.com.
Visit this group at http://groups.google.com/group/open-iscsi.
For more options, visit https://groups.google.com/d/optout.


Re: problems connecting to Equallogic 6100

2013-04-02 Thread Mark Lehrer



iscsiadm -m node --login


I have an EqualLogic and it works pretty well.

I typically specify the target with -T, and the grapdmin user console
shows errors too (if the initator is connect to a good IP).

Is it timing out or failing right away?

Can you double check that the target portal IP is correct?  Maybe try
defining the target as unrestricted to make sure it isn't a password
kind of problem.

Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to open-iscsi+unsubscr...@googlegroups.com.
To post to this group, send email to open-iscsi@googlegroups.com.
Visit this group at http://groups.google.com/group/open-iscsi?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.




Re: Multipath or not ?

2013-03-11 Thread Mark Lehrer



I have a virtual tape library and a iscsi SAN. All have multiple
ethernet interfaces, This will ressult in multiples sessions to the
targets.So I wonder if I must use dm-multipath or not ? Does the


Typically I only use multipath if I have one initiator talking to one
target.  Otherwise I just use 802.3ad/LACP and let the switch try to
figure out the best way to route things.

Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to open-iscsi+unsubscr...@googlegroups.com.
To post to this group, send email to open-iscsi@googlegroups.com.
Visit this group at http://groups.google.com/group/open-iscsi?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.




Re: Ubuntu server + open-iscsi + ocfs2. Connectivity loss causes immediate server reboot.

2012-06-14 Thread Mark Lehrer
storage HP P2000 G3 iscsi. Even short network connectivity loss is 
causing immediate server crash and reboot. In syslog I can not found any 


Does ocfs2 have a fencing arrangement like Red Hat's gfs ?  This is a 
feature actually, clustered filesystems don't want rogue machines writing 
straight to the block device.


If this is the problem, there should be a way to tweak the timeouts.  Red 
Hat's cluster system has an overwhelming amount of options available.


FWIW, we gave up on Red Hat Cluster  GFS, which is similar to OCFS, due to 
this type of outage.  We switched to ASM and NFS.  ASM works great with 
iscsi.


Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-iscsi@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.



Re: General question on shared storage

2011-05-17 Thread Mark Lehrer
AFAIK, Linux buffers blocks from block devices. So if host A reads a 
block from shared storage (e.g. the MBR), and later host B modifies 
that block, will host A still have the last block read in cache? If 
so there's a big problem.


I am only familiar with GFS and ASM, and both of those will typically 
disable any sort of caching.


 
Consider that you want to mount a non-clustered filesystem 
alternatingly on two hosts (like for a HA failover). If host A used 
the filesystem, the host B took over the filesystem without booting 


I would recommend only logging in from one host at a time.  Some SAN types 
(e.g. EqualLogic) will allow multiple hosts to discover the target, but will 
only allow one to be logged in.  This will help with split brain 
corruption issues.


In general, though, you should be careful of trying to force a SAN type of 
solution when you really need NAS.  NFS and Samba are quite mature and easy 
to deal with.


Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-iscsi@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.



Re: Changes made on iSCSI disk are not shown on initiator

2011-05-12 Thread Mark Lehrer

the mounted folder) I can see this file afterwards only on the
initiator, but not on the target. Means, I open a Terminal on the


The basic rule is that if you are going to export storage to a single 
machine, use iSCSI.  If you need shared storage, use NFS instead.


You could consider a shared storage file system like Oracle's ASM, but they 
aren't so easy to deal with.  NFS4 has locking, user authentication, and 
everything you need built into the protocol.


Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-iscsi@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.



Re: How to resurrect SCSI device

2011-03-02 Thread Mark Lehrer
I'm facing a somewhat strange situation. As a part of testing a 
multipath solution, I've written a script that simulates the failure and


This is an excellent thread, thanks.

Does this happen even without multipathd?  I have been trying to work on a 
procedure to fix a similar problem where the device is read-only even though 
the session was re-established - and there is no way to either unmount it or 
get it back to a state where it is usable.


Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-iscsi@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.



Re: iscsiadm: discovery record not found

2011-02-22 Thread Mark Lehrer

   iscsiadm: Discovery record [$DP,3260] not found.


It looks like you have a shell syntax problem here.  DP should be the IP 
address (aka portal).


Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-iscsi@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.



Re: iscsi and virtualization

2011-01-24 Thread Mark Lehrer

This is apparently not a good idea - since, as far as I can tell at
least, I have to restart the iscsi subsystem every time I present a
new disk from the PS6000. A service iscsi restart effectively
destroys all KVM guests using existing, already logged in iscsi disks
will hang indefinitely wondering where their disks disappeared to.


Ouch, you definitely don't want to do this.

You should be able to just do a discovery and find new targets.  The 
commands take some getting used to, though.  Check out section 5.2 in this: 
http://www.open-iscsi.org/docs/README


We have had pretty good luck with Equallogic  RHEL5/CentOS 5.

It would be a bit cleaner if there were a way to tell discovery of a 
particular target that you want to add, instead of doing a full scan, but it 
should be safe.


Mark

--
You received this message because you are subscribed to the Google Groups 
open-iscsi group.
To post to this group, send email to open-iscsi@googlegroups.com.
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.