|
Ninja Rope posted:Yes, but my point is that those companies aren't just downloading 10-BETA2 and putting it on production servers and calling it a day. They have FreeBSD committers on staff to work cherry-pick patches and test only specific configs, as well as fix bugs as they occur (and feed them back upstream). I didn't understand the question to be "is FreeBSD suitable for production", which in a general sense it is, but "is the latest FreeBSD 10 beta suitable for a random internet user to use in production", which I think the answer is no. It will be, but it's not right now. To be honest, this workflow is just as applicable to Linux. The people who are doing serious performance work find odd bugs, fix or report them upstream, and cherry pick patches to backport onto older kernels (especially for large clusters). Nothing about this is unique to BSD, and I'd be just as comfortable running 10-BETA or Debian unstable or Fedora 19 for an nginx+Sinatra+passenger or postgres. You run RHEL or SLES for stable ABIs for enterprise software which needs them (Clearcase, OracleDB, whatever), and APIs to a lesser degree (though BSD is much better about this), not necessarily performance or stability under load.
|
# ¿ Nov 7, 2013 04:56 |
|
|
# ¿ May 14, 2024 22:29 |
|
lostleaf posted:Did anyone else have problems with freebsd-update? I'm upgrading from 9.1-release to 10.0-release and it keeps on reporting integrity check fail. A quick google search tells me that I have to upgrade to the latest 9.x release then upgrade from that to 10.0. Is it really supposed to be that draconian? Yes, it is.
|
# ¿ Jan 21, 2014 16:36 |
|
Riso posted:As a client, doesn't it just pretend to be a NT4? It's the opposite, really. As a server, it pretends to be an NT4-style domain (except for Samba4, which peers with 2012R2). winbind will happily talk Kerberos.
|
# ¿ Feb 10, 2014 22:33 |
|
roadhead posted:Messing around upgrading some ports, mostly using 'portupgrade -a' and now I get this Because "import SABnzbd" won't call __main__. Try reinstalling cffi
|
# ¿ Apr 15, 2014 17:58 |
|
roadhead posted:I reinstalled that port already. What this backtrace is telling is: sab is trying to start cherrypy. Cherrypy is using openssl through pycrypto pycrypto is using terrible/dangerous bindings. Those bindings are calling out to a native C library. You have a few possibilities: cffi is broken, but you updated it. openssl was updated by pycrypto was not pycrypto was updated by openssl was not Other than cffi, it's likely that you got the heartbleed fix, and security/py-pycrypto needs to be rebuilt against the new library version.
|
# ¿ Apr 15, 2014 18:29 |
|
roadhead posted:Ok I did a "make reinstall clean" for py-pycrypto and openssl and still no go. Is there something else I should try besides that command? What are the versions of cffi and pycrypto?
|
# ¿ Apr 15, 2014 22:16 |
|
roadhead posted:py27-cffi-0.8.2 You'll probably have to ask the cffi port maintainer.
|
# ¿ Apr 15, 2014 23:20 |
|
Ninja Rope posted:Unrelated, but what is a better python openssl API wrapper to use/why is that terrible? Essentially, see this. There's nothing inherently wrong with it if you need to access OpenSSL or CommonCrypto primitives directly, but it's potentially dangerous. This is broadly true of everything involving SSL, though, and there's not really a better wrapper. They're all equally bad.
|
# ¿ Apr 16, 2014 19:35 |
|
thebigcow posted:^^ Wouldn't NFS be easier between BSD and OSX? Or are they not quite compatible and require a bunch of work? NFS is just terrible on OSX.
|
# ¿ Jun 6, 2014 20:12 |
|
sarehu posted:Last time I tried that (shortly after 9.1(?) was released) it didn't work, there were no ports available. sarehu posted:
Please look at the confs in /etc/pkg/ and /usr/local/etc/pkg.conf It sounds like your repository config is broken.
|
# ¿ Nov 24, 2014 06:20 |
|
wooger posted:OK, so I posted a while back about my problems with PCBSD: Namely, I couldn't get the Nvidia driver working at all, and hence run Gnome-Shell. Just portmaster -a I don't have any problems with vlc from pkgng, at least. What's wrong with gnome-shell? Install it, enable hal and dbus, install nvidia-drivers and enable them in loader.conf, start gnome-shell
|
# ¿ Dec 5, 2014 15:29 |
|
wooger posted:Portmaster -a will remove all pkgng packages I have installed? No. Portmaster -a will look for upgrades to installed packages, via ports. Adding that to rc.conf should be fine. What's happening when you try to start gnome?
|
# ¿ Dec 5, 2014 16:54 |
|
Xenomorph posted:Well, after wondering why my Logwatch settings never seemed to work right, I just found out the correct path for the configuration file on FreeBSD isn't mentioned anywhere. It's not a bug. The docs may be a bug, because they seem to have just packaged the default documentation, but as a general rule: if the binary is in /usr/local (because you installed it from ports or packages), the config file is in /usr/local, as is the init script
|
# ¿ May 5, 2015 18:28 |
|
D. Ebdrup posted:I've been playing around with bhyve a lot, but didn't know this. Since you posted, I've been playing around with it and it's actually really neat. The netstack gets used everywhere, but that's not a great indication of usage. Windows also used it for a long time. JunOS is also pulled from comparatively ancient versions of FreeBSD, even recent versions. FreeBSD leaves a lot to be desired in realtime applications, but there's a realtime patchset and canbus drivers, so it's possible, but nobody's talking about it if so, without even getting into certifying it to work in given scenarios. QNX and VxWorks are still very much kings of that (embedded realtime OS) hill.
|
# ¿ Oct 8, 2015 19:35 |
|
|
# ¿ May 14, 2024 22:29 |
|
If you want questionable advice, use launchd
|
# ¿ Apr 7, 2017 06:03 |