Re: [OpenIndiana-discuss] [solarisx86] For those nostalgic fellas
har har har! it is a three hump camel for sure. Sent from Jasons' hand held On Aug 6, 2012, at 9:15 PM, Alan Coopersmith wrote: > Holy water ... check. > Wooden stakes ... check. > Silver bullets ... check. > Chainsaw ... check. > > Okay, all ready here, we'll make sure it can't get back in. > >-alan- > > On 08/ 6/12 08:47 PM, Jerry Kemp wrote: >> reposted from the Sun rescue mailing list. >> >> Jerry >> >> >> . >> So. >> >> Is no one going to comment on the new CDE news? >> >> http://cdesktopenv.sourceforge.net/ >> >> It has now been open sourced under the LGPL. >> >> More info here: >> http://sourceforge.net/p/cdesktopenv/wiki/Home/ >> . >> >> On 01/ 2/11 12:15 PM, Alan Coopersmith wrote: >>> On 01/ 2/11 05:55 AM, Apostolos Syropoulos wrote: http://devio.us/~kpedersen/ >>> >>> Nooo! I just killed CDE! It can't be back from the dead! >>> Run before the zombie gets us! >>> >>> [Unfortunately, since CDE is a closed gate, the bug data isn't published >>> on bugs.opensolaris.org, so I can't link to the bug report for the "fix" >>> in snv_156 that deleted 12937 source files for the CDE applications, >>> leaving just the libraries like Motif for binary compatibility. >>> Unfortunately, OpenMotif isn't binary compatible with the Motif 2.1 in >>> Solaris.] >>> > > > -- >-Alan Coopersmith- alan.coopersm...@oracle.com > Oracle Solaris Engineering - http://blogs.oracle.com/alanc > > ___ > OpenIndiana-discuss mailing list > OpenIndiana-discuss@openindiana.org > http://openindiana.org/mailman/listinfo/openindiana-discuss ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] TCP Reset Packet Problem
Do you have the VBox client utilities installed? I had seen strange clock problems when the Agent is not installed. Mike On Tue, 2012-08-07 at 12:31 +0800, Patrick Yu wrote: > I was actually trying to say "strange problem of TCP reset packet (or > the the lack of)". :-) > > Anyway, after some more hours of digging around, I found some leads: > > # ndd tcp tcp_rst_sent_rate_enabled > 1 > # ndd tcp tcp_rst_sent_rate > 40 > # kstat tcp 1 1 | egrep '[Rr]st' > outRsts 874 > tcp_rst_unsent 3644 > # telnet 127.0.0.1 12345 > Trying 127.0.0.1... > ^C > # kstat tcp 1 1 | egrep '[Rr]st' > outRsts 875 > tcp_rst_unsent 3648 > > The rst sent rate of 40 a second seems not being observed, despite > there's no reset packets generated in the system except for the test > run. I did some more tests: When trying to increase the rst_sent_rate, > it takes a value of 800+ to make reset packets work, and the value > needs to be further incremented when more reset packets are being > sent. It seems like the counter for reset packets per second never get > zeroed. > > Looks like a real bug to me. But I am still not sure how to trigger > this - it runs fine in the first day of two before exhibiting this > strange behavior. I even did some "stress" test from > https://blogs.oracle.com/clive/entry/tcp_reset_delay to a freshly > rebooted system in failed attempts to reproduce the erroneous > conditions. But I am sure it will come back when it's left there for > another day. > > I suspect it could be the time accuracy problem due to it being a vbox > VM. I looked at tcp_output.c from > https://hg.openindiana.org/upstream/illumos/illumos-gate/file/adffc698eaf5/usr/src/uts/common/inet/tcp/tcp_output.c#l3279 > and tried to change the clock backwards and forwards, but still could > not reproduce it. > > Now, my temporary workaround is to set 0 to tcp_rst_sent_rate_enabled, > but in effect totally disable any tcp reset DOS protection. Hope this > could help someone with a similar case. > > Best regards, > Patrick > > On Mon, Aug 6, 2012 at 3:09 PM, Patrick Yu wrote: > > Hi, > > > > I am experiencing a very strange TCP problem (the lack of) with my new > > oi_151a5 install. The machine ran fine on the first day or two after a > > fresh reboot, and after that SSH connections broke down and hanged > > mysteriously during SSL handshake where no connections could be made > > from both outside or even from inside using loopback lo0. > > > > It took me awhile to track it down to this bug - > > https://www.illumos.org/issues/1983 where the workaround posted solved > > my SSH problem. But upon closer examination I found the source of the > > problem is actually something else in my particular case. It turns out > > any TCP connections to a closed port that is not being listened to > > would not generate a TCP reset packet from the networking core. Any > > clients connecting to these ports would hang there indefinitely for > > lengthy retries. > > > > I initially thought it was due to ipfilter but even after I cleared > > the table, RST was still not being sent no matter what interface was > > involved (lo0, e1000g0). The connection and RST packet would come back > > after a reboot, and the problem recurs after a few days even with > > low/no load as this is a testing installation running as a VM. > > > > Things like X didn't start properly when there's missing TCP RST. I > > didn't have time to look into it, but I presume it's related to this > > problem too. Worth nothing is that those ports being listened to > > exhibited no problems whatsoever - I can even do a iperf across the > > network with very good results. > > > > I could do some silly thing like the below ipf.conf snippet to "force" > > RST packet being sent. But then if there's any pass statement at the > > end like "pass in quick on lo0", RST would disappear again! > > set intercept_loopback true; > > block return-rst in > > > > Anyone has an idea what could be the cause? A misconfiguration or a > > bug? Any pointer would be greatly appreciated. I still keep a snapshot > > of the problematic VM and am ready to do some more experiments with > > it. Below is what the problematic session looks like, and a normal > > snoop after reboot. > > > > # telnet 127.0.0.1 12345 > > Trying 127.0.0.1... > > ^C > > > > # snoop -I lo0 -tr -r > > Using device ipnet/lo0 (promiscuous mode) > > 0.0127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > > Seq=1227588634 Len=0 Win=32768 Options= > 71716119 0,nop,wscale 2> > > 1.13752127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > > Seq=1227588634 Len=0 Win=32768 Options= > 71716119 0,nop,wscale 2> > > 3.40631127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > > Seq=1227588634 Len=0 Win=32768 Options= > 71716119 0,nop,wscale 2> > > 7.92479127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > > Seq=1227588634 Len=
Re: [OpenIndiana-discuss] TCP Reset Packet Problem
I was actually trying to say "strange problem of TCP reset packet (or the the lack of)". :-) Anyway, after some more hours of digging around, I found some leads: # ndd tcp tcp_rst_sent_rate_enabled 1 # ndd tcp tcp_rst_sent_rate 40 # kstat tcp 1 1 | egrep '[Rr]st' outRsts 874 tcp_rst_unsent 3644 # telnet 127.0.0.1 12345 Trying 127.0.0.1... ^C # kstat tcp 1 1 | egrep '[Rr]st' outRsts 875 tcp_rst_unsent 3648 The rst sent rate of 40 a second seems not being observed, despite there's no reset packets generated in the system except for the test run. I did some more tests: When trying to increase the rst_sent_rate, it takes a value of 800+ to make reset packets work, and the value needs to be further incremented when more reset packets are being sent. It seems like the counter for reset packets per second never get zeroed. Looks like a real bug to me. But I am still not sure how to trigger this - it runs fine in the first day of two before exhibiting this strange behavior. I even did some "stress" test from https://blogs.oracle.com/clive/entry/tcp_reset_delay to a freshly rebooted system in failed attempts to reproduce the erroneous conditions. But I am sure it will come back when it's left there for another day. I suspect it could be the time accuracy problem due to it being a vbox VM. I looked at tcp_output.c from https://hg.openindiana.org/upstream/illumos/illumos-gate/file/adffc698eaf5/usr/src/uts/common/inet/tcp/tcp_output.c#l3279 and tried to change the clock backwards and forwards, but still could not reproduce it. Now, my temporary workaround is to set 0 to tcp_rst_sent_rate_enabled, but in effect totally disable any tcp reset DOS protection. Hope this could help someone with a similar case. Best regards, Patrick On Mon, Aug 6, 2012 at 3:09 PM, Patrick Yu wrote: > Hi, > > I am experiencing a very strange TCP problem (the lack of) with my new > oi_151a5 install. The machine ran fine on the first day or two after a > fresh reboot, and after that SSH connections broke down and hanged > mysteriously during SSL handshake where no connections could be made > from both outside or even from inside using loopback lo0. > > It took me awhile to track it down to this bug - > https://www.illumos.org/issues/1983 where the workaround posted solved > my SSH problem. But upon closer examination I found the source of the > problem is actually something else in my particular case. It turns out > any TCP connections to a closed port that is not being listened to > would not generate a TCP reset packet from the networking core. Any > clients connecting to these ports would hang there indefinitely for > lengthy retries. > > I initially thought it was due to ipfilter but even after I cleared > the table, RST was still not being sent no matter what interface was > involved (lo0, e1000g0). The connection and RST packet would come back > after a reboot, and the problem recurs after a few days even with > low/no load as this is a testing installation running as a VM. > > Things like X didn't start properly when there's missing TCP RST. I > didn't have time to look into it, but I presume it's related to this > problem too. Worth nothing is that those ports being listened to > exhibited no problems whatsoever - I can even do a iperf across the > network with very good results. > > I could do some silly thing like the below ipf.conf snippet to "force" > RST packet being sent. But then if there's any pass statement at the > end like "pass in quick on lo0", RST would disappear again! > set intercept_loopback true; > block return-rst in > > Anyone has an idea what could be the cause? A misconfiguration or a > bug? Any pointer would be greatly appreciated. I still keep a snapshot > of the problematic VM and am ready to do some more experiments with > it. Below is what the problematic session looks like, and a normal > snoop after reboot. > > # telnet 127.0.0.1 12345 > Trying 127.0.0.1... > ^C > > # snoop -I lo0 -tr -r > Using device ipnet/lo0 (promiscuous mode) > 0.0127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > Seq=1227588634 Len=0 Win=32768 Options= 71716119 0,nop,wscale 2> > 1.13752127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > Seq=1227588634 Len=0 Win=32768 Options= 71716119 0,nop,wscale 2> > 3.40631127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > Seq=1227588634 Len=0 Win=32768 Options= 71716119 0,nop,wscale 2> > 7.92479127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > Seq=1227588634 Len=0 Win=32768 Options= 71716119 0,nop,wscale 2> > 16.93940127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn > Seq=1227588634 Len=0 Win=32768 Options= 71716119 0,nop,wscale 2> > ^C# > > # ifconfig lo0 > lo0: flags=2001000849 mtu > 8232 index 1 > inet 127.0.0.1 netmask ff00 > # > # netstat -r -n | grep lo0 > 127.0.0.1127.0.0.1UH5 9638 lo0
Re: [OpenIndiana-discuss] [solarisx86] For those nostalgic fellas
Holy water ... check. Wooden stakes ... check. Silver bullets ... check. Chainsaw ... check. Okay, all ready here, we'll make sure it can't get back in. -alan- On 08/ 6/12 08:47 PM, Jerry Kemp wrote: > reposted from the Sun rescue mailing list. > > Jerry > > > . > So. > > Is no one going to comment on the new CDE news? > > http://cdesktopenv.sourceforge.net/ > > It has now been open sourced under the LGPL. > > More info here: > http://sourceforge.net/p/cdesktopenv/wiki/Home/ > . > > On 01/ 2/11 12:15 PM, Alan Coopersmith wrote: >> On 01/ 2/11 05:55 AM, Apostolos Syropoulos wrote: >>> http://devio.us/~kpedersen/ >> >> Nooo! I just killed CDE! It can't be back from the dead! >> Run before the zombie gets us! >> >> [Unfortunately, since CDE is a closed gate, the bug data isn't published >> on bugs.opensolaris.org, so I can't link to the bug report for the "fix" >> in snv_156 that deleted 12937 source files for the CDE applications, >> leaving just the libraries like Motif for binary compatibility. >> Unfortunately, OpenMotif isn't binary compatible with the Motif 2.1 in >> Solaris.] >> -- -Alan Coopersmith- alan.coopersm...@oracle.com Oracle Solaris Engineering - http://blogs.oracle.com/alanc ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] [solarisx86] For those nostalgic fellas
reposted from the Sun rescue mailing list. Jerry . So. Is no one going to comment on the new CDE news? http://cdesktopenv.sourceforge.net/ It has now been open sourced under the LGPL. More info here: http://sourceforge.net/p/cdesktopenv/wiki/Home/ . On 01/ 2/11 12:15 PM, Alan Coopersmith wrote: > On 01/ 2/11 05:55 AM, Apostolos Syropoulos wrote: >> http://devio.us/~kpedersen/ > > Nooo! I just killed CDE! It can't be back from the dead! > Run before the zombie gets us! > > [Unfortunately, since CDE is a closed gate, the bug data isn't published > on bugs.opensolaris.org, so I can't link to the bug report for the "fix" > in snv_156 that deleted 12937 source files for the CDE applications, > leaving just the libraries like Motif for binary compatibility. > Unfortunately, OpenMotif isn't binary compatible with the Motif 2.1 in > Solaris.] > ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] ActiveDirectory UID mapping (netatalk)
Hi, I've got a server hooked up to a 2003 AD and CIFS and netatalk are both allowing AD users to login (netatalk 3 via PAM). One thing that's a bit puzzling is that the afpd process correctly gets the correct username mapping (and shows up as being owned by the correct user with a ps listing), but whatever the user writes is only written as UID 60001 (ie. nobody). Has anyone got netatalk (or anything in fact) to use the same mapping that the CIFS service uses? I assumed at first that this may not be possible, but the fact that the actual afpd process does seem to be run under the correct UID seems to suggest that the mapping is at least going on. Thanks, James. ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] Installing TrueType Fonts
On 08/ 6/12 02:56 PM, Mark Creamer wrote: > I managed to figure this out once in the past, but unfortunately did not > make my usual copious notes and now I can't seem to get it to work on a new > server. The issue is that I want to install true-type fonts, as available > from the links on this page http://corefonts.sourceforge.net/ so my > Jasper reporting application that's running on OI 151_a4 can use them. > > I was able to download the webfonts.tar.gz package, un-tar/gunzip it, which > leaves me with a group of Microsoft executables, one for each font. I then > used cabextract running on a Linux box to extract all the TTFs from the > Microsoft exe. I then copied the fonts into /etc/fonts on the OI box, and > ran fc-cache /etc/fonts. That finishes without giving me an error, but the > developer of the report tells me the font is still not working. /etc/fonts is for configuration files. The actual fonts should be under a directory like /usr/share/fonts that is listed in one of the configuration files font path entries. -- -Alan Coopersmith- alan.coopersm...@oracle.com Oracle Solaris Engineering - http://blogs.oracle.com/alanc ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] Installing TrueType Fonts
I managed to figure this out once in the past, but unfortunately did not make my usual copious notes and now I can't seem to get it to work on a new server. The issue is that I want to install true-type fonts, as available from the links on this page http://corefonts.sourceforge.net/ so my Jasper reporting application that's running on OI 151_a4 can use them. I was able to download the webfonts.tar.gz package, un-tar/gunzip it, which leaves me with a group of Microsoft executables, one for each font. I then used cabextract running on a Linux box to extract all the TTFs from the Microsoft exe. I then copied the fonts into /etc/fonts on the OI box, and ran fc-cache /etc/fonts. That finishes without giving me an error, but the developer of the report tells me the font is still not working. If anyone has done this before, can you give me a tip on what I might be missing? Then I can document it this time :-) Thanks! -- Mark ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On 08/06/2012 09:19 PM, Richard Elling wrote: > On Aug 6, 2012, at 5:15 AM, James Carlson wrote: >> >> It's never been possible to mount NFS at boot. > > Well, some of us old farts remember nd, and later, NFS-based diskless > workstations :-) > The current lack of support for diskless leaves an empty feeling in my heart > :-P > -- richard I don't think of myself as an old fart and I remember and love diskless. As technology trends rotate around in about a 10-15 year cycle, what goes around, comes around. Diskless will be the new cloud! Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On Aug 6, 2012, at 5:15 AM, James Carlson wrote: > > It's never been possible to mount NFS at boot. Well, some of us old farts remember nd, and later, NFS-based diskless workstations :-) The current lack of support for diskless leaves an empty feeling in my heart :-P -- richard -- ZFS Performance and Training richard.ell...@richardelling.com +1-760-896-4422 ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] OpenSolaris on SPARC
Hi Martin, hate to pester you but I just got my two 1.2 ghz chips in my sb1000 and was trying to put an old version of opensolaris on it (to hold me over till OI) and it can't find drivers for any of the video cards during install. I have tried xvr1200,xvr100, and an old expert/elite 3d (not sure which). Is this to be expected? I have never tried to put open solaris on it before. Dan On 07/30/12 06:51 PM, Martin Bochnig wrote: On Mon, Jul 30, 2012 at 6:33 PM, Daniel Kjar wrote: Fantastic! I can't wait to fire up my sb1000 again! The SB1000/2000 is my favorite workstation, too ... as I have always said: http://mail.opensolaris.org/pipermail/xwin-discuss/2008-August/002281.html http://unix.derkeiler.com/Newsgroups/comp.sys.sun.hardware/2005-05/0232.html These days you get the x7310a (1.2GHz III Cu) module for pennies. And they do outperform a Dualcore SB2500 Red 1,28GHz. Some people now think, that the Blade 1000/2000 is an electricity waster (which is true to some degree, just make your own tests and decide). But ironically the only other Sun SPARC Workstations that manage to outperform the maxed out SB1000/2000 are Dual core 1.6GHz SB2500 Silver and Ultra 45, which consume the same amount, if I can believe these results: http://chrysalis.rutgers.edu/hardware/powerbrief.php And I read various reviews that confirm, that CAD users were disappointed by the IIIi based machines, that did not deliver any significant "feelable" performance boost over the SB1000/2000, despite their higher clock, on-die (though much smaller) cache and faster memory. A look to spec confirms this: CFP2000 Results: SB1000/2000 IIIcu @ 1.2 GHz: http://www.spec.org/cpu2000/results/res2003q2/cpu2000-20030326-02001.html SB2500Red IIIi @ 1.28 GHz http://www.spec.org/cpu2000/results/res2003q4/cpu2000-20030821-02436.html SB2500Silver IIIi @ 1.6 GHz http://www.spec.org/cpu2000/results/res2005q1/cpu2000-20050107-03681.html I have an xvr1200 but I have pretty much determined I will never be able to use that card again Why? The XVR-1200 is similar to a XVR-600. I know, under Xorg it would be never supported. Maybe the best one could expect would be unaccelerated fb mode. However, as we now have an opensrc Xsun and redistributable binary drivers, including redistributable /dev/fb drivers for initializing and mapping the card, plus my changes to make the stumbled / cut-down opened version of Xsun function again, nothing should prevent you from using your XVR-1200. And in contrast to previous versions of modified Xorg, if now doesn't matter anymore, into which system with what kind of pci bridge you intend to plug your card. It should function, period. I know, this sounds like a dream. And that's why we cannot ever thank Alan Coopersmith enough, that he pushed through the opensrc release of Xsun, after it had been EOL'ed in 2010, yet just moments before Oracle closed the tap :) Do you know what card would work best? I have an ati radeon (xvr 100?) card and a couple of those monstrous creator3ds. Any of these cards will work. The Creator/Creator3D/Elite3D even function with Xorg's ported Linux/BSD drivers under Xorg, that are still available on martux.org: http://martux.org/RELEASES/FOX1.0/SPARC/ http://martux.org/RELEASES/SPARC_distro_incl_Xorg7.2/ But under Xsun every of these just functions, does so reliably and quickly. So it rather depends on your 2D vs. 3D needs, free PCI slots vs. UPA slots, power consumption and so on. I in my beloved SB2000 use an Elite 3D UPA card. Maybe the XVR-1000 (UPA) or 1200 (3D Labs Wildcat PCI) would be even faster. But for normal office needs, any of these should be more than sufficient. rgds. %mab ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss -- Dr. Daniel Kjar Assistant Professor of Biology Division of Mathematics and Natural Sciences Elmira College 1 Park Place Elmira, NY 14901 607-735-1826 http://faculty.elmira.edu/dkjar "...humans send their young men to war; ants send their old ladies" -E. O. Wilson ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] rpool on mirrored USB flash drives
On 08/06/2012 04:43 PM, Joshua M. Clulow wrote: > On 7 August 2012 00:21, Sašo Kiselkov wrote: >> I'm considering rebuilding my home server (an HP MicroServer) for some >> higher capacity and dumping the single 250GB rpool disk and instead >> replace it with a pair of USB flash sticks (no need to preserve data, >> I'll copy stuff around and help the system along with booting). The >> question is, is this a good way to go about it? >> Are there any gotchas with using USB attached storage as boot media in >> OI? Any experience with that? (For all I know, it's smooth and working.) > > I have tried several different USB sticks inside my HP Microserver as > the rpool for OI installs. While the system supports doing it the > experience was universally poor; ranging from pathologically slow > boots right down to periodic checksum errors on some sticks under > heavy I/O load to rpool filesystems. > > I ended up installing a 60GB SSD in the optical drive bay and > attaching it to the fifth internal SATA port (the "OD" port), having > applied a BIOS hack to turn it into an AHCI port. This worked very > well. Thanks, this clears it up for me. Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] rpool on mirrored USB flash drives
On 7 August 2012 00:21, Sašo Kiselkov wrote: > I'm considering rebuilding my home server (an HP MicroServer) for some > higher capacity and dumping the single 250GB rpool disk and instead > replace it with a pair of USB flash sticks (no need to preserve data, > I'll copy stuff around and help the system along with booting). The > question is, is this a good way to go about it? > Are there any gotchas with using USB attached storage as boot media in > OI? Any experience with that? (For all I know, it's smooth and working.) I have tried several different USB sticks inside my HP Microserver as the rpool for OI installs. While the system supports doing it the experience was universally poor; ranging from pathologically slow boots right down to periodic checksum errors on some sticks under heavy I/O load to rpool filesystems. I ended up installing a 60GB SSD in the optical drive bay and attaching it to the fifth internal SATA port (the "OD" port), having applied a BIOS hack to turn it into an AHCI port. This worked very well. I would only recommend USB sticks for booting a SmartOS-like distribution where you do hardly any writing to the boot device (ever) and only read around 250MB from it into RAM *once* per boot. -- Joshua M. Clulow UNIX Admin/Developer http://blog.sysmgr.org ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] rpool on mirrored USB flash drives
On 06/08/2012 16:21, Sašo Kiselkov wrote: I'm considering rebuilding my home server (an HP MicroServer) for some higher capacity and dumping the single 250GB rpool disk and instead replace it with a pair of USB flash sticks (no need to preserve data, I'll copy stuff around and help the system along with booting). The question is, is this a good way to go about it? ... Are there any gotchas with using USB attached storage as boot media in OI? Any experience with that? (For all I know, it's smooth and working.) It's actually an option for some Sun servers (X4550, X6275), but personally I won't do that if you need swap,dump and other stuff constantly changing (databases, etc.). Instead, on my home workstation I run everything from SSDs now (no spinning rust at all), and that really flies. Never saw THAT small boot times before (BIOS-Firmware takes longer than OI boot), and responsiveness on commands is A LOT better. But, of course, that's still a bit expensive... -- Dr.Udo GrabowskiInst.f.Meteorology a.Climate Research IMK-ASF-SAT www-imk.fzk.de/asf/sat/grabowski/ www.imk-asf.kit.edu/english/sat.php KIT - Karlsruhe Institute of Technologyhttp://www.kit.edu Postfach 3640,76021 Karlsruhe,Germany T:(+49)721 608-26026 F:-926026 ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
the first box I had was sol8 but it was inherited and apparently 'fixed' if this has been around for longer than that. I did move to sol from linux so I may just be confusing my first automount experience. On 08/ 6/12 10:12 AM, Michael Stapleton wrote: Hi, I have been using Solaris only since 2.6 and /home has always been an autofs mount point. What version of Solaris did not have /home as an autofs mount point? Or are you confusing OI with some other OS? Mike . On Mon, 2012-08-06 at 09:51 -0400, Daniel Kjar wrote: I am sure it has great value, why would it be there otherwise?. I just remember installing a fresh version solaris one day and trying to figure out why I couldn't just delete export and use /home like always. Therefore, in my mind I associate it with being frustrated by a "I am sorry Dave, I can't allow you to do that" message until I figured out what had changed. On 08/ 6/12 09:47 AM, James Carlson wrote: Daniel Kjar wrote: Really? What do you call that crap in etc under auto_master and auto_home? Read the man pages for the automounter. Start with automount(1M). Yes, the system comes by default with that "crap," but (a) you certainly are under no obligation to use /export/home if you don't like it and (b) the mechanism that underlies it is far more general than just auto_home. It allows you to trigger configured mounts based on file system access, and handles fail-over, platform-related variable expansion, and directory service integration. ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss -- Dr. Daniel Kjar Assistant Professor of Biology Division of Mathematics and Natural Sciences Elmira College 1 Park Place Elmira, NY 14901 607-735-1826 http://faculty.elmira.edu/dkjar "...humans send their young men to war; ants send their old ladies" -E. O. Wilson ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] rpool on mirrored USB flash drives
I did try that when building my current home server with just one USB stick. My experience was that the resulting OS is quite a bit slower, may be due to a bad choice of stick on my part. Anyway, I ended up with 1x500GB rpool disk and 5x2TB RAID-Z1 pool using an Asus mATX motherboard and i3-2120T. Alex. On Mon, Aug 6, 2012 at 3:21 PM, Sašo Kiselkov wrote: > I'm considering rebuilding my home server (an HP MicroServer) for some > higher capacity and dumping the single 250GB rpool disk and instead > replace it with a pair of USB flash sticks (no need to preserve data, > I'll copy stuff around and help the system along with booting). The > question is, is this a good way to go about it? I plan on using the USB > sticks only for holding rpool and booting off of them. All other data > will be on a RAID-z consisting of 4x 2TB disks, so speed isn't all that > necessary. I'm somewhat constrained by the number of SATA ports > available, but I have plenty of USB slots (1 internal and 6 external) > where I could put the media (and since the box is hidden away in a > closet, nobody is ever going to be touching it). > > Are there any gotchas with using USB attached storage as boot media in > OI? Any experience with that? (For all I know, it's smooth and working.) > > Cheers, > -- > Saso > > ___ > OpenIndiana-discuss mailing list > OpenIndiana-discuss@openindiana.org > http://openindiana.org/mailman/listinfo/openindiana-discuss ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] rpool on mirrored USB flash drives
I'm considering rebuilding my home server (an HP MicroServer) for some higher capacity and dumping the single 250GB rpool disk and instead replace it with a pair of USB flash sticks (no need to preserve data, I'll copy stuff around and help the system along with booting). The question is, is this a good way to go about it? I plan on using the USB sticks only for holding rpool and booting off of them. All other data will be on a RAID-z consisting of 4x 2TB disks, so speed isn't all that necessary. I'm somewhat constrained by the number of SATA ports available, but I have plenty of USB slots (1 internal and 6 external) where I could put the media (and since the box is hidden away in a closet, nobody is ever going to be touching it). Are there any gotchas with using USB attached storage as boot media in OI? Any experience with that? (For all I know, it's smooth and working.) Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Hi, I have been using Solaris only since 2.6 and /home has always been an autofs mount point. What version of Solaris did not have /home as an autofs mount point? Or are you confusing OI with some other OS? Mike . On Mon, 2012-08-06 at 09:51 -0400, Daniel Kjar wrote: > I am sure it has great value, why would it be there otherwise?. I just > remember installing a fresh version solaris one day and trying to > figure out why I couldn't just delete export and use /home like always. > > Therefore, in my mind I associate it with being frustrated by a "I am > sorry Dave, I can't allow you to do that" message until I figured out > what had changed. > > On 08/ 6/12 09:47 AM, James Carlson wrote: > > Daniel Kjar wrote: > >> Really? What do you call that crap in etc under auto_master and auto_home? > > Read the man pages for the automounter. Start with automount(1M). > > > > Yes, the system comes by default with that "crap," but (a) you certainly > > are under no obligation to use /export/home if you don't like it and (b) > > the mechanism that underlies it is far more general than just auto_home. > > It allows you to trigger configured mounts based on file system access, > > and handles fail-over, platform-related variable expansion, and > > directory service integration. > > > ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
I am sure it has great value, why would it be there otherwise?. I just remember installing a fresh version solaris one day and trying to figure out why I couldn't just delete export and use /home like always. Therefore, in my mind I associate it with being frustrated by a "I am sorry Dave, I can't allow you to do that" message until I figured out what had changed. On 08/ 6/12 09:47 AM, James Carlson wrote: Daniel Kjar wrote: Really? What do you call that crap in etc under auto_master and auto_home? Read the man pages for the automounter. Start with automount(1M). Yes, the system comes by default with that "crap," but (a) you certainly are under no obligation to use /export/home if you don't like it and (b) the mechanism that underlies it is far more general than just auto_home. It allows you to trigger configured mounts based on file system access, and handles fail-over, platform-related variable expansion, and directory service integration. -- Dr. Daniel Kjar Assistant Professor of Biology Division of Mathematics and Natural Sciences Elmira College 1 Park Place Elmira, NY 14901 607-735-1826 http://faculty.elmira.edu/dkjar "...humans send their young men to war; ants send their old ladies" -E. O. Wilson ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Daniel Kjar wrote: > Really? What do you call that crap in etc under auto_master and auto_home? Read the man pages for the automounter. Start with automount(1M). Yes, the system comes by default with that "crap," but (a) you certainly are under no obligation to use /export/home if you don't like it and (b) the mechanism that underlies it is far more general than just auto_home. It allows you to trigger configured mounts based on file system access, and handles fail-over, platform-related variable expansion, and directory service integration. -- James Carlson 42.703N 71.076W ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On Mon, Aug 6, 2012 at 3:33 PM, Daniel Kjar wrote: > Really? What do you call that crap in etc under auto_master and auto_home? ah ... "every elephant is an animal, but not every animal is an elephant" ;-) (in other words: there's other applications of the automounter than mounting peoples' home directories) regards Michael > On 08/ 6/12 09:31 AM, James Carlson wrote: >> >> Daniel Kjar wrote: >>> >>> I would never use a remote etc. Lose the network and the box becomes >>> unusable. Not good. That automounter drives me batty. I hate the >>> whole export/home thing and remove the auto_home crap and reboot as soon >>> as I set up a new server. >> >> automounter != export/home >> > > -- > Dr. Daniel Kjar > Assistant Professor of Biology > Division of Mathematics and Natural Sciences > Elmira College > 1 Park Place > Elmira, NY 14901 > 607-735-1826 > http://faculty.elmira.edu/dkjar > > "...humans send their young men to war; ants send their old ladies" > -E. O. Wilson > > > > > > ___ > OpenIndiana-discuss mailing list > OpenIndiana-discuss@openindiana.org > http://openindiana.org/mailman/listinfo/openindiana-discuss -- Michael Schuster http://recursiveramblings.wordpress.com/ ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Really? What do you call that crap in etc under auto_master and auto_home? On 08/ 6/12 09:31 AM, James Carlson wrote: Daniel Kjar wrote: I would never use a remote etc. Lose the network and the box becomes unusable. Not good. That automounter drives me batty. I hate the whole export/home thing and remove the auto_home crap and reboot as soon as I set up a new server. automounter != export/home -- Dr. Daniel Kjar Assistant Professor of Biology Division of Mathematics and Natural Sciences Elmira College 1 Park Place Elmira, NY 14901 607-735-1826 http://faculty.elmira.edu/dkjar "...humans send their young men to war; ants send their old ladies" -E. O. Wilson ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Daniel Kjar wrote: > I would never use a remote etc. Lose the network and the box becomes > unusable. Not good. That automounter drives me batty. I hate the > whole export/home thing and remove the auto_home crap and reboot as soon > as I set up a new server. automounter != export/home -- James Carlson 42.703N 71.076W ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On 08/06/2012 03:21 PM, James Carlson wrote: > Daniel Kjar wrote: >> Thanks for asking about this. I had just assumed the same. Machines I >> updated all the way from opensolaris still mounted nfs at boot but my >> clean 151a5s did not. Very annoying when apache fails cause it didn't >> mount htdocs from the file server. > > OK. It's possible that I'm just wrong about this. The only case where > I've seen NFS entries in /etc/vfstab was with mount-at-boot set to "no." > (And since the introduction of ZFS, I don't really put much in vfstab > anymore.) > > In any event, I think automount provides an easier-to-manage solution. > And unlike some other operating systems, it actually works reliably on > OpenIndiana. > > But regardless of mounting mechanism, I don't think I'd consider using > NFS for anything under /etc. It sort of defeats the point of that > directory to have anything remote there. Even if I'm sharing > configuration between machines, I set up rsync to keep them up-to-date, > so that a failure (or unreachability) of the NFS server doesn't cause > otherwise-independent machines to fail. While I agree with your advice in general, it doesn't in my case. Rest assured, I have considered the possibility of a network failure and using NFS is entirely suitable for my needs here. This is a network streaming service, so if the network is down, the fact that it can't start up due to an unavailable NFS mount is the least of my problems. Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On 06/08/2012 15:18, Michael Schuster wrote: On Mon, Aug 6, 2012 at 3:04 PM, Sašo Kiselkov wrote: On 08/06/2012 02:15 PM, James Carlson wrote: Sašo Kiselkov wrote: I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails to automount on an OI client at boot, even though the filesystem is clearly marked as "mount at boot": #device device mount FS fsckmount mount #to mount to fsck point typepassat boot options 192.168.133.1:/etc/streamers- /etc/streamers nfs - yes nodevices,nosetuid,ro When I issue /sbin/mountall or mount -a, it mounts just fine. Needless to say, its failure to mount at boot results in failure of dependent services to start, which is quite bummer if I have to do it manually each time. Nothing is logged to /var/adm/messages, so I have no idea why it ignores my NFS mounts at boot. It simply does. Anybody got an idea on how to track this down? It's never been possible to mount NFS at boot. Well, apparent it has, since it's working after I enabled some NFS client services. I'd say you're lucky - I wouldn't rely on this working every time. If svc:nfs/client is active, it does a 'mountall -F nfs' after milestone/network has been fired up, so that always works. -- Dr.Udo GrabowskiInst.f.Meteorology a.Climate Research IMK-ASF-SAT www-imk.fzk.de/asf/sat/grabowski/ www.imk-asf.kit.edu/english/sat.php KIT - Karlsruhe Institute of Technologyhttp://www.kit.edu Postfach 3640,76021 Karlsruhe,Germany T:(+49)721 608-26026 F:-926026 ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
I would never use a remote etc. Lose the network and the box becomes unusable. Not good. That automounter drives me batty. I hate the whole export/home thing and remove the auto_home crap and reboot as soon as I set up a new server. On 08/ 6/12 09:21 AM, James Carlson wrote: Daniel Kjar wrote: Thanks for asking about this. I had just assumed the same. Machines I updated all the way from opensolaris still mounted nfs at boot but my clean 151a5s did not. Very annoying when apache fails cause it didn't mount htdocs from the file server. OK. It's possible that I'm just wrong about this. The only case where I've seen NFS entries in /etc/vfstab was with mount-at-boot set to "no." (And since the introduction of ZFS, I don't really put much in vfstab anymore.) In any event, I think automount provides an easier-to-manage solution. And unlike some other operating systems, it actually works reliably on OpenIndiana. But regardless of mounting mechanism, I don't think I'd consider using NFS for anything under /etc. It sort of defeats the point of that directory to have anything remote there. Even if I'm sharing configuration between machines, I set up rsync to keep them up-to-date, so that a failure (or unreachability) of the NFS server doesn't cause otherwise-independent machines to fail. -- Dr. Daniel Kjar Assistant Professor of Biology Division of Mathematics and Natural Sciences Elmira College 1 Park Place Elmira, NY 14901 607-735-1826 http://faculty.elmira.edu/dkjar "...humans send their young men to war; ants send their old ladies" -E. O. Wilson ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Daniel Kjar wrote: > Thanks for asking about this. I had just assumed the same. Machines I > updated all the way from opensolaris still mounted nfs at boot but my > clean 151a5s did not. Very annoying when apache fails cause it didn't > mount htdocs from the file server. OK. It's possible that I'm just wrong about this. The only case where I've seen NFS entries in /etc/vfstab was with mount-at-boot set to "no." (And since the introduction of ZFS, I don't really put much in vfstab anymore.) In any event, I think automount provides an easier-to-manage solution. And unlike some other operating systems, it actually works reliably on OpenIndiana. But regardless of mounting mechanism, I don't think I'd consider using NFS for anything under /etc. It sort of defeats the point of that directory to have anything remote there. Even if I'm sharing configuration between machines, I set up rsync to keep them up-to-date, so that a failure (or unreachability) of the NFS server doesn't cause otherwise-independent machines to fail. -- James Carlson 42.703N 71.076W ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On Mon, Aug 6, 2012 at 3:04 PM, Sašo Kiselkov wrote: > On 08/06/2012 02:15 PM, James Carlson wrote: >> Sašo Kiselkov wrote: >>> I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails >>> to automount on an OI client at boot, even though the filesystem is >>> clearly marked as "mount at boot": >>> >>> #device device mount FS fsckmount >>> mount >>> #to mount to fsck point typepassat boot >>> options >>> 192.168.133.1:/etc/streamers- /etc/streamers nfs - >>> yes nodevices,nosetuid,ro >>> >>> When I issue /sbin/mountall or mount -a, it mounts just fine. Needless >>> to say, its failure to mount at boot results in failure of dependent >>> services to start, which is quite bummer if I have to do it manually >>> each time. Nothing is logged to /var/adm/messages, so I have no idea why >>> it ignores my NFS mounts at boot. It simply does. Anybody got an idea on >>> how to track this down? >> >> It's never been possible to mount NFS at boot. > > Well, apparent it has, since it's working after I enabled some NFS > client services. I'd say you're lucky - I wouldn't rely on this working every time. Have you investigated using the automounter? regards Michael -- Michael Schuster http://recursiveramblings.wordpress.com/ ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Thanks for asking about this. I had just assumed the same. Machines I updated all the way from opensolaris still mounted nfs at boot but my clean 151a5s did not. Very annoying when apache fails cause it didn't mount htdocs from the file server. On 08/ 6/12 09:04 AM, Sašo Kiselkov wrote: On 08/06/2012 02:15 PM, James Carlson wrote: Sašo Kiselkov wrote: I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails to automount on an OI client at boot, even though the filesystem is clearly marked as "mount at boot": #device device mount FS fsckmount mount #to mount to fsck point typepassat boot options 192.168.133.1:/etc/streamers- /etc/streamers nfs - yes nodevices,nosetuid,ro When I issue /sbin/mountall or mount -a, it mounts just fine. Needless to say, its failure to mount at boot results in failure of dependent services to start, which is quite bummer if I have to do it manually each time. Nothing is logged to /var/adm/messages, so I have no idea why it ignores my NFS mounts at boot. It simply does. Anybody got an idea on how to track this down? It's never been possible to mount NFS at boot. Well, apparent it has, since it's working after I enabled some NFS client services. Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss -- Dr. Daniel Kjar Assistant Professor of Biology Division of Mathematics and Natural Sciences Elmira College 1 Park Place Elmira, NY 14901 607-735-1826 http://faculty.elmira.edu/dkjar "...humans send their young men to war; ants send their old ladies" -E. O. Wilson ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On 08/06/2012 02:15 PM, James Carlson wrote: > Sašo Kiselkov wrote: >> I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails >> to automount on an OI client at boot, even though the filesystem is >> clearly marked as "mount at boot": >> >> #device device mount FS fsckmount >> mount >> #to mount to fsck point typepassat boot >> options >> 192.168.133.1:/etc/streamers- /etc/streamers nfs - >> yes nodevices,nosetuid,ro >> >> When I issue /sbin/mountall or mount -a, it mounts just fine. Needless >> to say, its failure to mount at boot results in failure of dependent >> services to start, which is quite bummer if I have to do it manually >> each time. Nothing is logged to /var/adm/messages, so I have no idea why >> it ignores my NFS mounts at boot. It simply does. Anybody got an idea on >> how to track this down? > > It's never been possible to mount NFS at boot. Well, apparent it has, since it's working after I enabled some NFS client services. Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
On 08/06/2012 02:03 PM, Udo Grabowski (IMK) wrote: > For nfsv3 > svc:/network/nfs/client:default > should be enabled. > > All these should be enabled for nfsv4 > online Jun_29 svc:/network/nfs/cbd:default > online Jun_29 svc:/network/nfs/status:default > online Jun_29 svc:/network/nfs/mapid:default > online Jun_29 svc:/network/nfs/rquota:default > online Jun_29 svc:/network/nfs/nlockmgr:default > > And, for automount, enable > svc:/system/filesystem/autofs:default A great many thanks, enabling these services solved it. Cheers, -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
Sašo Kiselkov wrote: > I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails > to automount on an OI client at boot, even though the filesystem is > clearly marked as "mount at boot": > > #device device mount FS fsckmount > mount > #to mount to fsck point typepassat boot > options > 192.168.133.1:/etc/streamers- /etc/streamers nfs - > yes nodevices,nosetuid,ro > > When I issue /sbin/mountall or mount -a, it mounts just fine. Needless > to say, its failure to mount at boot results in failure of dependent > services to start, which is quite bummer if I have to do it manually > each time. Nothing is logged to /var/adm/messages, so I have no idea why > it ignores my NFS mounts at boot. It simply does. Anybody got an idea on > how to track this down? It's never been possible to mount NFS at boot. The normal "mount -a" takes place well before any of the networking infrastructure -- interfaces, routes, and the like -- are available. Making it work at boot poses a chicken-and-egg problem. You need to have the local file systems mounted in order to start up the networking services ... but if the mount process itself requires networking, you're stuck. This is exactly what the automounter is for ... though I somewhat doubt that it makes adminstrative sense to have anything remote-mounted under /etc. The whole point of /etc is that it's local. -- James Carlson 42.703N 71.076W ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
Re: [OpenIndiana-discuss] NFS fails to automount at boot
For nfsv3 svc:/network/nfs/client:default should be enabled. All these should be enabled for nfsv4 online Jun_29 svc:/network/nfs/cbd:default online Jun_29 svc:/network/nfs/status:default online Jun_29 svc:/network/nfs/mapid:default online Jun_29 svc:/network/nfs/rquota:default online Jun_29 svc:/network/nfs/nlockmgr:default And, for automount, enable svc:/system/filesystem/autofs:default On 06/08/2012 13:56, Sašo Kiselkov wrote: I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails to automount on an OI client at boot, even though the filesystem is clearly marked as "mount at boot": #device device mount FS fsckmount mount #to mount to fsck point typepassat boot options 192.168.133.1:/etc/streamers- /etc/streamers nfs - yes nodevices,nosetuid,ro When I issue /sbin/mountall or mount -a, it mounts just fine. Needless to say, its failure to mount at boot results in failure of dependent services to start, which is quite bummer if I have to do it manually each time. Nothing is logged to /var/adm/messages, so I have no idea why it ignores my NFS mounts at boot. It simply does. Anybody got an idea on how to track this down? -- Dr.Udo GrabowskiInst.f.Meteorology a.Climate Research IMK-ASF-SAT www-imk.fzk.de/asf/sat/grabowski/ www.imk-asf.kit.edu/english/sat.php KIT - Karlsruhe Institute of Technologyhttp://www.kit.edu Postfach 3640,76021 Karlsruhe,Germany T:(+49)721 608-26026 F:-926026 ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] NFS fails to automount at boot
I've run into a bizzare issue. An NFS export I have in /etc/vfstab fails to automount on an OI client at boot, even though the filesystem is clearly marked as "mount at boot": #device device mount FS fsckmount mount #to mount to fsck point typepassat boot options 192.168.133.1:/etc/streamers- /etc/streamers nfs - yes nodevices,nosetuid,ro When I issue /sbin/mountall or mount -a, it mounts just fine. Needless to say, its failure to mount at boot results in failure of dependent services to start, which is quite bummer if I have to do it manually each time. Nothing is logged to /var/adm/messages, so I have no idea why it ignores my NFS mounts at boot. It simply does. Anybody got an idea on how to track this down? -- Saso ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
[OpenIndiana-discuss] TCP Reset Packet Problem
Hi, I am experiencing a very strange TCP problem (the lack of) with my new oi_151a5 install. The machine ran fine on the first day or two after a fresh reboot, and after that SSH connections broke down and hanged mysteriously during SSL handshake where no connections could be made from both outside or even from inside using loopback lo0. It took me awhile to track it down to this bug - https://www.illumos.org/issues/1983 where the workaround posted solved my SSH problem. But upon closer examination I found the source of the problem is actually something else in my particular case. It turns out any TCP connections to a closed port that is not being listened to would not generate a TCP reset packet from the networking core. Any clients connecting to these ports would hang there indefinitely for lengthy retries. I initially thought it was due to ipfilter but even after I cleared the table, RST was still not being sent no matter what interface was involved (lo0, e1000g0). The connection and RST packet would come back after a reboot, and the problem recurs after a few days even with low/no load as this is a testing installation running as a VM. Things like X didn't start properly when there's missing TCP RST. I didn't have time to look into it, but I presume it's related to this problem too. Worth nothing is that those ports being listened to exhibited no problems whatsoever - I can even do a iperf across the network with very good results. I could do some silly thing like the below ipf.conf snippet to "force" RST packet being sent. But then if there's any pass statement at the end like "pass in quick on lo0", RST would disappear again! set intercept_loopback true; block return-rst in Anyone has an idea what could be the cause? A misconfiguration or a bug? Any pointer would be greatly appreciated. I still keep a snapshot of the problematic VM and am ready to do some more experiments with it. Below is what the problematic session looks like, and a normal snoop after reboot. # telnet 127.0.0.1 12345 Trying 127.0.0.1... ^C # snoop -I lo0 -tr -r Using device ipnet/lo0 (promiscuous mode) 0.0127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn Seq=1227588634 Len=0 Win=32768 Options= 1.13752127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn Seq=1227588634 Len=0 Win=32768 Options= 3.40631127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn Seq=1227588634 Len=0 Win=32768 Options= 7.92479127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn Seq=1227588634 Len=0 Win=32768 Options= 16.93940127.0.0.1 -> 127.0.0.1TCP D=12345 S=36692 Syn Seq=1227588634 Len=0 Win=32768 Options= ^C# # ifconfig lo0 lo0: flags=2001000849 mtu 8232 index 1 inet 127.0.0.1 netmask ff00 # # netstat -r -n | grep lo0 127.0.0.1127.0.0.1UH5 9638 lo0 ::1 ::1 UH 71612 lo0 # # ipf -Fa # # ipfstat -io empty list for ipfilter(out) empty list for ipfilter(in) # # netstat -anv | grep 12345 # # svccfg -s ipfilter:default listprop |grep firewall_config firewall_config_default com.sun,fw_configuration firewall_config_default/value_authorization astring solaris.smf.value.firewall.config firewall_config_default/version count1 firewall_config_default/apply_to astring firewall_config_default/exceptionsastring firewall_config_default/policyastring custom firewall_config_default/custom_policy_fileastring /etc/ipf/ipf.conf firewall_config_default/open_portsastring firewall_config_override com.sun,fw_configuration firewall_config_override/apply_to astring firewall_config_override/value_authorization astring solaris.smf.value.firewall.config firewall_config_override/policy astring none # # reboot # # telnet 127.0.0.1 12345 Trying 127.0.0.1... telnet: Unable to connect to remote host: Connection refused # # snoop -I lo0 -tr -r Using device ipnet/lo0 (promiscuous mode) 0.0127.0.0.1 -> 127.0.0.1TCP D=12345 S=53940 Syn Seq=1084268217 Len=0 Win=32768 Options= 0.5127.0.0.1 -> 127.0.0.1TCP D=53940 S=12345 Rst Ack=1084268218 Win=0 ^C# Thanks. Best regards, Patrick ___ OpenIndiana-discuss mailing list OpenIndiana-discuss@openindiana.org http://openindiana.org/mailman/listinfo/openindiana-discuss
