This is a multi-part message in MIME format. --------------68695C298F307D3C48D2ABAE Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 8bit Hi, As we are using a very similar hardware and usage as Mark (Dell poweredge hosts, Dell Equallogic SAN, iSCSI, and tons of LUNs for all those VMs), I'm jumping into this thread. Le 12/01/2017 à 16:29, Yaniv Kaul a écrit :
While it's a bit of a religious war on what is preferred with iSCSI - network level bonding (LACP) or multipathing on the iSCSI level, I'm on the multipathing side. The main reason is that you may end up easily using just one of the paths in a bond - if your policy is not set correct on how to distribute connections between the physical links (remember that each connection sticks to a single physical link. So it really depends on the hash policy and even then - not so sure). With iSCSI multipathing you have more control - and it can also be determined by queue depth, etc. (In your example, if you have SRC A -> DST 1 and SRC B -> DST 1 (as you seem to have), both connections may end up on the same physical NIC.)
If we reduce the number of storage domains, we reduce the number of devices and therefore the number of LVM Physical volumes that appear in Linux correct? At the moment each connection results in a Linux device which has its own queue. We have some guests with high IO loads on their device whilst others are low. All the storage domain / datastore sizing guides we found seem to imply its a trade-off between ease of management (i.e not having millions of domains to manage), IO contention between guests on a single large storage domain / datastore and possible wasted space on storage domains. If you have further information on recommendations, I am more than willing to change things as this problem is making our environment somewhat unusable at the moment. I have hosts that I cant bring online and therefore reduced resiliency in clusters. They used to work just fine but the environment has grown over the last year and we also upgraded the Ovirt version from 3.6 to 4.x. We certainly had other problems, but host activation wasnt one of them and its a problem thats driving me mad.
I would say that each path has its own device (and therefore its own queue). So I'd argue that you may want to have (for example) 4 paths to each LUN or perhaps more (8?). For example, with 2 NICs, each connecting to two controllers, each controller having 2 NICs (so no SPOF and nice number of paths).
Here, one key point I'm trying (to no avail) to discuss for years with Redhat people, and either I did not understood, either I wasn't clear enough, or Redhat people answered me they owned no Equallogic SAN to test it, is : My (and maybe many others) Equallogic SAN has two controllers, but is publishing only *ONE* virtual ip address. On one of our other EMC SAN, publishing *TWO* ip addresses, which can be published in two different subnets, I fully understand the benefits and working of multipathing (and even in the same subnet, our oVirt setup is happily using multipath). But on one of our oVirt setup using the Equallogic SAN, we have no choice but point our hosts iSCSI interfaces to one single SAN ip, so no multipath here. At this point, we saw no other mean than using bonding mode 1 to reach our SAN, which is terrible for storage experts. To come back to Mark's story, we are still using 3.6.5 DCs and planning to upgrade. Reading all this is making me delay this step. -- Nicolas ECARNOT --------------68695C298F307D3C48D2ABAE Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: 8bit <html> <head> <meta content="text/html; charset=windows-1252" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <div class="moz-cite-prefix">Hi,<br> <br> As we are using a very similar hardware and usage as Mark (Dell poweredge hosts, Dell Equallogic SAN, iSCSI, and tons of LUNs for all those VMs), I'm jumping into this thread.<br> <br> Le 12/01/2017 à 16:29, Yaniv Kaul a écrit :<br> </div> <blockquote cite="mid:CAJgorsZaZOk6yaqH7mQH5bXRkegvLN_3pmGZ6J53TEwXtdnmvQ@mail.gmail.com" type="cite"> <div dir="ltr"> <div class="gmail_extra"> <div class="gmail_quote"><br> <div>While it's a bit of a religious war on what is preferred with iSCSI - network level bonding (LACP) or multipathing on the iSCSI level, I'm on the multipathing side. The main reason is that you may end up easily using just one of the paths in a bond - if your policy is not set correct on how to distribute connections between the physical links (remember that each connection sticks to a single physical link. So it really depends on the hash policy and even then - not so sure). With iSCSI multipathing you have more control - and it can also be determined by queue depth, etc.</div> <div>(In your example, if you have SRC A -> DST 1 and SRC B -> DST 1 (as you seem to have), both connections may end up on the same physical NIC.)</div> <div><br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div link="blue" vlink="purple" lang="EN-GB"> <div class="m_7631641247395573674WordSection1"> <p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif"></span></p> <p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif"> </span></p> <p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif">If we reduce the number of storage domains, we reduce the number of devices and therefore the number of LVM Physical volumes that appear in Linux correct? At the moment each connection results in a Linux device which has its own queue. We have some guests with high IO loads on their device whilst others are low. All the storage domain / datastore sizing guides we found seem to imply its a trade-off between ease of management (i.e not having millions of domains to manage), IO contention between guests on a single large storage domain / datastore and possible wasted space on storage domains. If you have further information on recommendations, I am more than willing to change things as this problem is making our environment somewhat unusable at the moment. I have hosts that I cant bring online and therefore reduced resiliency in clusters. They used to work just fine but the environment has grown over the last year and we also upgraded the Ovirt version from 3.6 to 4.x. We certainly had other problems, but host activation wasnt one of them and its a problem thats driving me mad.</span></p> </div> </div> </blockquote> <div><br> </div> <div>I would say that each path has its own device (and therefore its own queue). So I'd argue that you may want to have (for example) 4 paths to each LUN or perhaps more (8?). For example, with 2 NICs, each connecting to two controllers, each controller having 2 NICs (so no SPOF and nice number of paths).</div> </div> </div> </div> </blockquote> Here, one key point I'm trying (to no avail) to discuss for years with Redhat people, and either I did not understood, either I wasn't clear enough, or Redhat people answered me they owned no Equallogic SAN to test it, is :<br> My (and maybe many others) Equallogic SAN has two controllers, but is publishing only *ONE* virtual ip address.<br> On one of our other EMC SAN, publishing *TWO* ip addresses, which can be published in two different subnets, I fully understand the benefits and working of multipathing (and even in the same subnet, our oVirt setup is happily using multipath).<br> <br> But on one of our oVirt setup using the Equallogic SAN, we have no choice but point our hosts iSCSI interfaces to one single SAN ip, so no multipath here.<br> <br> At this point, we saw no other mean than using bonding mode 1 to reach our SAN, which is terrible for storage experts.<br> <br> <br> To come back to Mark's story, we are still using 3.6.5 DCs and planning to upgrade.<br> Reading all this is making me delay this step.<br> <br> -- <br> Nicolas ECARNOT<br> </body> </html> --------------68695C298F307D3C48D2ABAE--