Hello,

I will add just one more:

- uniting all the forks across illumos, so sharing/testing of innovations is 
possible on larger scale and defragment the ecosystem becausr it helps noone.

Cheers,
Adam

> On 14 Sep 2017, at 20:05, Pasi Kärkkäinen <[email protected]> wrote:
> 
>> On Thu, Sep 14, 2017 at 04:49:26PM +0000, Garrett D'Amore wrote:
>>   illumos *had* Xen support.  It may still be there and work?  I don't know
>>   if anyone has done anything with it in ages.
>> 
> 
> Afaik DilOS has patches for supporting more recent versions of Xen than XVM 
> in Illumos/Opensolaris.
> 
> http://www.dilos.org/
> https://dilos-dev.atlassian.net/wiki/spaces/DS/pages/1343539/Xen
> 
> 
> -- Pasi
> 
>>   On Thu, Sep 14, 2017 at 9:06 AM Miles Fidelman
>>   <[1][email protected]> wrote:
>> 
>>>     On 9/14/17 3:30 AM, Perttu Aaltonen wrote:
>>> 
>>> Hello all. I've been following the list for a long time and rarely
>>     post anything, but now wanted to chime in.
>>>> On 14 Sep 2017, at 10:31, Alexander Pyhalov <[2][email protected]> wrote:
>>>> 
>>>> 2) Zone/VM live migration. Don't tell me "it's too hard, use
>>     clusters". I want to have an illumos-based cluster for virtualization
>>     purposes, VMware/Linux KVM/even OpenVZ do it, why can't we?
>>> I agree. There are a lot of "legacy" software that are not engineered
>>     for the cloud and only runs on Windows. The developers of these kinds of
>>     products seem to often think clustering/HA as a solved problem with
>>     VMware and not bother building it to the product itself. But VMware
>>     replication/HA/Fault Tolerance is complex, finicky and often out of
>>     reach for very small businesses due to cost. I'm not very familiar with
>>     HAST and DRBD, but seems like the added complexity might sometimes
>>     outweigh the expected benefits there as well. I've read varying success
>>     with OmniOS as a built in storage provider for ESXi through iSCSI or
>>     NFS.
>>> 
>>> Ideally this would be built in to Illumos/ZFS. I already use SmartOS
>>     and Znapzend for replicating KVM VM's between hosts, but it's only crash
>>     consistent. VSS or file system quiescing support would improve this. In
>>     small business space, there usually isn't a need for a full lock-step
>>     fault tolerance type automated HA with shared storage and what have you,
>>     so ability to bring up replicated VM's on a backup host in a consistent
>>     state would go a long way. I've just recently set up replication between
>>     ESXi hosts with Veeam and ZFS send would just be so much faster and
>>     easier to set up and use.
>>> 
>>> Better yet, LX Zone type ability to run Windows software directly in a
>>     zone would be a dream, but probably quite impossible.
>>> 
>>> 
>>     Well how about just porting Xen & DRBD to Illumos.  Not so much
>>     "innovation" as simply basic porting.  (Now, DRBD is 2-node only, so
>>     maybe Sheepdog - again, not innovation, but incredibly useful.)
>> 
>>     Miles Fidelman
>> 
>>     --
>>     In theory, there is no difference between theory and practice.
>>     In practice, there is.  .... Yogi Berra
>> 

------------------------------------------
illumos-discuss
Archives: 
https://illumos.topicbox.com/groups/discuss/discussions/T83f198c8597cf8e3-M65a5609ef802d0977697f208
Powered by Topicbox: https://topicbox.com

Reply via email to