FreeNAS 9.2.0-RC2 now available

[pbucher]

No joy on trying to load the vmxnet3 ESXi drivers. I updated my FreeNAS build box to FreeBSD 9.2 and it still works with the prebuilt binary drivers. I used the box to build the vmxnet3 driver from source and verified it's operation under my newly updated FreeBSD 9.2. I then moved the 9.2 driver over to my FreeNAS box and still no go.

So was was changed between FreeBSD 9.2 & the TrueOS version used to build FreeNAS 9.2 that is causing the symbol _mtx_lock_flags to be undefined? I'll file a bug report on this also.

 

[Dusan]

If you have a FreeNAS build box why don't you compile the matching .ko yourself?

 

[pbucher]

Mostly because I need to figure out how to cross compile the vmxnet3 driver. The box is FreeBSD 9.2 so it will make a FreeBSD driver which won't work. I've hacked up the box build with the headers from the correct TrueOS release now and I'm testing the driver.

edit: If anyone needs a vmxnet3 driver(built from sources included with 5.5) for 9.2 PM me.

 

[jkh]

that right there, thats what scares the living shit out of me.

If you'd prefer I not be honest with you guys, I can edit my posts to be more marketing friendly. The tone of your reply is, in fact, merely a way of guaranteeing that - it adds no value to the discussion or in any way encourages "the right behavior", you're just preaching. I detest preachers. They are what lead to political correctness trumping all else in communciation and I, for one, resent it enough that you just wound up in my *plonk* file. :)

 

[gpsguy]

Leave out the marketing hype and keep the honesty.

I am grateful that the CTO is a participant on this forum.


Sent from my phone

 

[Magnus33]

There are bugs which is to be expected since we are not talking a massive company with loads of people and even those companies can put out buggy software.

The one thing i noticed about freenas is that system destroying bugs aren't going to make it through into the releases but you can expect the performance and useability ones to pop up.
There just no way to avoid some cropping up and i much rather see these bugs then ones that are going to cause my hard-drives to shoot out of the case and fly around the room looking for prey. :)

I myself had a native block error pop up which is hardly the end of the world and autotune didn't import correctly with saved config but that was easily fixable.
The none native block error came from a Hitachi hard-drive which i been meaning to replace anyway with a already bought drive.

9.2 still needs some work like real time hardware monitoring but performance is up and its running very solid

 

[Magnus33]

If you'd prefer I not be honest with you guys, I can edit my posts to be more marketing friendly. The tone of your reply is, in fact, merely a way of guaranteeing that - it adds no value to the discussion or in any way encourages "the right behavior", you're just preaching. I detest preachers. They are what lead to political correctness trumping all else in communication and I, for one, resent it enough that you just wound up in my *plonk* file. :)

Stick with honest please ;)
We may not always agree but much prefer the truth to BS any day.
Some always forget this is free software and if they chose to update there setup to a RC or beta they do so at their own risk.

The desire to have the latest and greatest should never override common sense when you got a lot to loose.
As always we are responsible for our own actions and should never try to place the blame on others..more so when we are getting something totally free.

 

[ThreeDee]

I for one .. as well as my family .. really appreciate all the hard work that goes into these releases. Thanks ... the update from 9.1 has gone smoothly (aside from some user error on my part) ..and the plex update is going just as smooth .. very cool stuff

 

[Scareh]

If you'd prefer I not be honest with you guys, I can edit my posts to be more marketing friendly. The tone of your reply is, in fact, merely a way of guaranteeing that - it adds no value to the discussion or in any way encourages "the right behavior", you're just preaching. I detest preachers. They are what lead to political correctness trumping all else in communciation and I, for one, resent it enough that you just wound up in my *plonk* file. :)

I never said for one that I don't appreciate your honesty on the subject of QA. For a fact I love the fact that you even admit to cut corners in this RC. Doesn't mean it can't still scare the living shit out of me for an actual release.
And maybe I am/was preaching, don't know, don't care, I still like my data hence the need for me to repley to your statement.
I still like freenas, and trust you guys to do the right thing but the fact that you think me being worried and me voicing my opinion on that subject is reason enough for me to end up in the *plonk* file.
That right there, that says it all. And i'll leave it at that.

 

[jonnn]

If you'd prefer I not be honest with you guys, I can edit my posts to be more marketing friendly. The tone of your reply is, in fact, merely a way of guaranteeing that - it adds no value to the discussion or in any way encourages "the right behavior", you're just preaching. I detest preachers. They are what lead to political correctness trumping all else in communciation and I, for one, resent it enough that you just wound up in my *plonk* file. :)

Don't worry about. There will always be someone out there who will bitch about anything.

This is free software and AFA I am concerned, I accept all risk. You guys provide amazing support and work for a free product. Keep it up.

Some people have never worked in a real corporate environment and don't know that things aren't always ideal in the real world. We wouldn't be human otherwise.

 

[Enlightend]

"I accept all risk" as anyone should when using non release level software.

Any level of critisism is acceptable when bugs land into a release version, but even with this one we're still talking about Release Candidates (granted some of the bugs mentioned should only exist in a beta level), but still, it isn't release, so it's should always be considered experimental and be allowed to go nowhere near live or important data.

 

[jkh]

The 9.2.0-RC2 bits may be found here: http://cdn.freenas.org/9.2.0/RC2/

And now we have a nightly which has all but one bug fixed (and that one is mild enough, having to do specially with manually created jails, that we may elect not to fix it for 9.2.0):

Check out http://cdn.freenas.org/nightlies/2013-12-18/ - we're very close, folks!

I hope that it also goes without saying that this is a *nightly* - it's not a fully QC'd build, it's just for testing.

Thanks!

- Jordan

 

[DrKK]

This looks a lot better Jordan. :)

I was able to successfully install it the way a putative end-user would do it, and not run into problems or bugs...except one thing, that may or may not be a problem:

The DHCP offer is accepted at boot time, link goes up, the IP address is good, but then, after a 15-20 second pause, the console menu says that there's a network problem, and to manually set it. Then I go in there, answer "yes" to "reset the interface" (or whatever), it re-asks for the DHCP, gets the SAME answer it got before, but THIS time, it actually accepts it, and reports that the GUI can be accessed on such-and-so IP.

Now, this is in VMWare player, *BUT*, this did *NOT* happen in VMWare Player for 9.1.1, I'm sure of that. So I'm not sure what to make of that--it's a very minor "bug", or at least, weird interaction with the VM's faux-NIC that wasn't there before in the 9.1.1 release.

But other than that, everything looks decent to me.

 

[jkh]

The DHCP offer is accepted at boot time, link goes up, the IP address is good, but then, after a 15-20 second pause, the console menu says that there's a network problem, and to manually set it. Then I go in there, answer "yes" to "reset the interface" (or whatever), it re-asks for the DHCP, gets the SAME answer it got before, but THIS time, it actually accepts it, and reports that the GUI can be accessed on such-and-so IP.

Now, this is in VMWare player

I don't know much about VMWare player as I have never used it (I've always just used the full-on VMWare product, be it ESXi 5.1 /5.5 or VMWare Fusion), but under VMWare Fusion 6.0.2 it seems to work just fine - the interface comes up properly the first time with DHCP, the web interface is there and waiting properly, etc.

Could this be a bug in VMWare Player? FreeNAS 9.1.1 may have worked, but the goalposts have also moved - FreeNAS 9.2.0 is based on FreeBSD 9.2 and a lot of VMware-specifc changes have occurred in that branch, as have changes to the VMWare tools. I could easily see some older release of VMWare player (I don't even know how that's versioned) having concerns with those changes.

Can you try this on something a bit more full-fledged and current, like ESXi 5.5? Thanks.

 

[cyberjock]

On my bare metal I don't have a problem with it.

And on the latest version of Virtualbox it works fine for me.

 

[Hi-Liter]

I think this one should also at least be investigated before the final release (plugins not working any longer): https://bugs.freenas.org/issues/3718 - Currently there are 3 people in the forums reporting this issue.

 

[cyberjock]

I think this one should also at least be investigated before the final release (plugins not working any longer): https://bugs.freenas.org/issues/3718 - Currently there are 3 people in the forums reporting this issue.

In all seriousness, from reading the bug reports on Crashplan in the bugs.freenas.org system it looks like Crashplan just doesn't work 'as well' as many would have hoped. As one person pointed out in one of the Crashplan tickets, the linux jail support in 9.2 is a far better option when 9.2 is released than using a plugin.

 

[Hi-Liter]

In all seriousness, from reading the bug reports on Crashplan in the bugs.freenas.org system it looks like Crashplan just doesn't work 'as well' as many would have hoped. As one person pointed out in one of the Crashplan tickets, the linux jail support in 9.2 is a far better option when 9.2 is released than using a plugin.

At least for me, the problem is not limited to Crashplan, the same behavior can also be observed with other plugins, e.g. plex media server.

 

[Z300M]

At least for me, the problem is not limited to Crashplan, the same behavior can also be observed with other plugins, e.g. plex media server.

Speaking of plugins... I have Logitech Media Server (aka Squeezebox Server) running in FreeNAS 9.1.1-RELEASE -- but I had to install it manually. Will/might I have to reinstall it after I update to FreeNAS 9.2.0?

 

[raidflex]

At least for me, the problem is not limited to Crashplan, the same behavior can also be observed with other plugins, e.g. plex media server.

I have been seeing this same error also. I did add my default gateway to the jail's network configuration, this seemed to help.