Re: [discuss] Illumos as a router?

2025-08-28 Thread Dan McDonald
> On Aug 28, 2025, at 10:03 AM, extrowerk wrote: > > Hello Dan, > > thanks for your email. You're welcome. > I plan to share the same IP prefix between the floors. I got a routable IPv6 > prefix, but IPv4 is CGNAT. Yeah... I'm lucky. I have a single public IP (which I *PAY* for :( ) on V

Re: [discuss] Illumos as a router?

2025-08-28 Thread Dan McDonald
Background and biases: - I've been using a "router zone" in illumos for over a decade. In my house, with feeds to four floors (basement, 1st, 2nd, attic). - For 8 years I've been routing two different, mildly disjoint, networks. - I do NOT use illumos as a switch (multiple ports, but shared IP

Re: [discuss] How to bhyve

2025-07-14 Thread Dan McDonald
On Jul 14, 2025, at 5:22 PM, Marcel Telka wrote: > > Hi, > > I'm trying to explore bhyve for the first time and so I tried some basic > example from the bhyve(8) man page. Surprisingly, it does not work. > > # mkfile -n 1g image > # bhyve -c 2 -s 0,hostbridge -s 1,lpc -s 2,virtio-blk,image \ >

Re: [discuss] ZFS native encryption status

2025-05-15 Thread Dan McDonald
On May 15, 2025, at 11:50 AM, Icos wrote: > > That is already pretty amazing!. Thanks for this work, just knowing that > eventually I will be able works for me. > > In the meantime there is always the option to backup to another Illumos host > I guess > illumos / illumos-discuss / see discussi

Re: [discuss] About eGPU, hotplug and passthrough

2025-03-18 Thread Dan McDonald
On Mar 18, 2025, at 5:43 PM, Joel Carnat via illumos-discuss wrote: > > Hello, > > Does hotplug and passthrough work on illumos to connect an nvidia eGPU via > thunderbolt to a bhyve Linux or lx zone? I don't know if thunderbolt devices even work on illumos or not. If it's "just" another PC

Re: [discuss] Flushing ippools

2024-09-30 Thread Dan McDonald
On Sep 30, 2024, at 4:33 AM, Sad Clouds wrote: > > Solaris 11.3 ported OpenBSD pf packet filter and I've seen some performance > tests from FreeBSD devs where it supposedly has much better performance > compared to ipf. I wonder if Illumos devs have plans to do something > similar... That's a

Re: [discuss] bhyve guest Solaris 11.4 with e1000g interfcae has a high IP latency

2024-07-10 Thread Dan McDonald
On Jul 10, 2024, at 12:35 PM, Carsten Grzemba via illumos-discuss wrote: > > Network analyzer tells me, there is a problem with the header checksum: > 0x incorrect, should be ... (may be caused by "IP checksum offload") > Is this a problem with the e1000g driver or is it configurable? Depe

Re: [discuss] Compiler(s) used to build Illumos

2024-04-14 Thread Dan McDonald
Josh answered this more completely on the developer's list. Please followup there. Dan -- illumos: illumos-discuss Permalink: https://illumos.topicbox.com/groups/discuss/Tf608666c4c97a989-Mc23ed34bcc8ebf8ba5518007 Delivery options: https://illumos.topi

Re: [discuss] Compiler(s) used to build Illumos

2024-04-14 Thread Dan McDonald
On Apr 14, 2024, at 7:10 PM, dan.ko...@gmail.com wrote: > > Greetings! I would like to find out which compiler(s) are > ued (or have been used) to build Illumos for Intel x64. Right now, it's the illumos-patched gcc10. Source available at https://github.com/illumos/gcc - specifcally the il_1

Re: [discuss] Illumos on ARM?

2024-04-12 Thread Dan McDonald
Pardon top post, on the road. Illumos can boot off low power x86 systems like the Alder Lake N100 now that igc(4D) for I225/226 is in, and a good chunk of those systems use igc. I run an 8gb one at home for Intel CPU fix tests. Dan Sent from my iPhone (typos, autocorrect, and all) > On Apr 1

Re: [discuss] Illumos and GNU

2024-02-21 Thread Dan McDonald
> On Feb 21, 2024, at 3:13 PM, Predrag Zečević > wrote: > > I have these packages: > :; pkg list | grep gnu ALL of these are not from illumos itself, but from supplemental places that distros use. Not sure if this is OI or OmniOS, but both, as mentioned by Josh earlier, have GNU-sourced

Re: [discuss] How to enable "flow label" in IPv6?

2024-02-03 Thread Dan McDonald
On Feb 3, 2024, at 1:50 AM, Bill Sommerfeld via illumos-discuss wrote: > > On 2/2/24 21:41, Dan McDonald wrote: >>> On Feb 2, 2024, at 11:07 PM, Jesus Cea wrote: >>> >>> I am using IPv6 and I noticed that by default SmartOS uses a flow label of >>>

Re: [discuss] How to enable "flow label" in IPv6?

2024-02-02 Thread Dan McDonald
> On Feb 2, 2024, at 11:07 PM, Jesus Cea wrote: > > I am using IPv6 and I noticed that by default SmartOS uses a flow label of 0 > (no flow label) in its IPv6 datagrams. I wonder how could I enable the use of > flow labels in order to be more compatible with for example, load balancers > (th

Re: [discuss] [Help]: 14744 Add TCP_INFO socket option

2023-11-19 Thread Dan McDonald
One more thing Something like this belongs more on the develo...@lists.illumos.org list. Dan -- illumos: illumos-discuss Permalink: https://illumos.topicbox.com/groups/discuss/Tbd27adee24ea3a50-M975a47ff59f8878ba5e4fb47 Delivery options: https://illu

Re: [discuss] [Help]: 14744 Add TCP_INFO socket option

2023-11-19 Thread Dan McDonald
I would provide (using gerrit) an illumos-gate-only version of TCP_INFO for review & testing. You already have a SmartOS PR for the curious about the LX portions that ALSO includes the -gate portions. https://github.com/TritonDataCenter/illumos-joyent/pull/458 Thanks for this! Dan --

Re: [discuss] update Emulex FC HBA driver

2023-10-17 Thread Dan McDonald
On Oct 17, 2023, at 4:33 AM, Carsten Grzemba via illumos-discuss wrote: > > I found the solution recently. I had to implement the new buffer autovalid > feature in the emlxs driver. Please make sure we see this work over on developer@? Thanks, Dan --

Re: [discuss] Weird reboots and last entries

2023-06-09 Thread Dan McDonald
On Jun 9, 2023, at 9:12 AM, Jon Tibble via illumos-discuss wrote: > > Hi, > > I've got a stand-alone SmartOS/Triton hypervisor that's been having some > unscheduled, seemingly random, reboots. This feels more core though which is > why I'm raising it here rather than the SmartOS ML. > > Thi

[discuss] CVE-2023-31284 Buffer Overflow in /dev/net

2023-05-04 Thread Dan McDonald
students at the University of Queensland for reporting this issue to us, and to Dan McDonald for his work in fixing it. The illumos Security Team -- illumos: illumos-discuss Permalink: https://illumos.topicbox.com/groups/discuss/Tafc2b6c1cce6c366-M3e316f20c78fc87421

Re: [discuss] Lets Encrypt/ Certbot for Illumos/ OmniOS/ OI

2023-03-29 Thread Dan McDonald
On Mar 29, 2023, at 10:29 AM, Peter Tribble wrote: > > I've used dehydrated extensively to get certs from Letsencrypt on both OmniOS > and Tribblix. It's a simple shell script that basically just works (only > minor gripe is > that it assumes awk = gawk so set your path appropriately) And grep

Re: [discuss] User auditing and filesystem monitoring

2023-01-20 Thread Dan McDonald
On Jan 20, 2023, at 12:44 PM, g...@napp-it.org wrote: > > > Due the issue tracker 11873 smb user auditing is at 100% since a few months. > As I asume its not yet in OmniOS I made this tests in OpenIndiana updated to > newest. The question is now, is 11873 already in Illumos or have I missed >

[discuss] Fwd: [developer] Heads up - Intel ucode 20220510 disables AVX-512 on Alder Lake

2022-06-22 Thread Dan McDonald
FYI, please ping developer's list if you are affected. Dan Begin forwarded message: From: Dan McDonald mailto:dan...@mnx.io>> Subject: [developer] Heads up - Intel ucode 20220510 disables AVX-512 on Alder Lake Date: June 22, 2022 at 3:52:11 PM EDT To: illumos-developer ma

[discuss] SmartOS and Triton moving from Joyent to MNX

2022-04-11 Thread Dan McDonald
phasing out of full commercial support for SmartOS and Triton. 2.) Two SmartOS and Triton engineers: Brian Bennett and Dan McDonald ("we") will be moving from Joyent to MNX. What is more important about this news is what does not change: - Paying customers will still have the

[discuss] SECURITY HEADS UP - illumos#14424

2022-01-18 Thread Dan McDonald
rived at the illumos fix. Thank you, Dan McDonald & Robert Mustacchi - on behalf of secur...@illumos.org -- illumos: illumos-discuss Permalink: https://illumos.topicbox.com/groups/discuss/T1c9e4f27f8c2f959-M8d6d96c6f3abf24bf40a7977 Delivery options: http

Re: [discuss] Broadcom HBA 9500 tri-mode support

2021-10-28 Thread Dan McDonald
According to its specifications tab, it uses a SAS3808 chipset. I don't see that mentioned even in the FreeBSD source, which according to the same specifications tab this board supports. I'm not sure what changes the SAS3808 brings to the table (perhaps NVMe?!?), and I'm also not sure how curr

Re: [discuss] unknown brand: "pkgsrc" ?

2021-08-02 Thread Dan McDonald
Jorge is right, I believe. OmniOS, last I checked, kept LX and other brands as distinct packages to be added by pkg(1). Dan -- illumos: illumos-discuss Permalink: https://illumos.topicbox.com/groups/discuss/Tba5f23fd6b9b2b09-M2aff092ea777b22e2dcf1a4b De

[discuss] FLAG DAY for kernel modules that call installctx()

2021-07-02 Thread Dan McDonald
Hello developers and developer-adjacents! With the impending integration of illumos#13915, the private-but-used-by-others installctx() call gains a 9th parameter, which is either NULL for old-behavior, or a non-NULL result of the new installctx_preallocate() call for installctx() callers which

[discuss] Re: [developer] Illumos/OI development system installation

2021-04-10 Thread Dan McDonald
On Apr 10, 2021, at 10:34 AM, Reginald Beardsley via illumos-developer wrote: > > I should like to suggest that providing a turnkey development installation > that would permit someone like me to compile format(1m) and fix the SEGV > faults would go a long way towards increasing the number o

Re: [discuss] AMD Epyc/Ryzen motherboard recommendations?

2020-05-13 Thread Dan McDonald
On May 12, 2020, at 10:33 AM, Robert Mustacchi wrote: > > I've been using the ASRock Rack EPYCD8 for some development work. So far > it's fine. The only thing I've found missing (and should probably look > at) is direct ipmi access on localhost. That speaks to the brand, which is good. Anyone i

[discuss] FLAG DAY: awk(1) is now nawk(1), oawk(1) is just oawk(1)

2020-04-21 Thread Dan McDonald
If this commit is a breaking change to anyone's scripts or software, oawk is still available and although it's now documented as obsolete, it will not be removed immediately. If you have any questions, please direct them to the developer's list (I've set Reply-To t

[discuss] HEADS UP: 11083 support NFS server in-zone

2020-02-07 Thread Dan McDonald
Hello folks! With the integration of illumos#11083 (support NFS server in-zone), aka IPD 11, illumos users may now run NFS servers inside a non-global zone. The illumos issue 12278 covers man page changes for this project, and will eventually contain items mentioned here. Distribution maintainer

Re: [discuss] 11083 NFS server in-zone

2020-01-29 Thread Dan McDonald
> On Jan 29, 2020, at 3:24 PM, Schweiss, Chip wrote: > > This looks exciting. Thanks (and thank Nexenta for the big wad of work). > One question, can zones be joined to different AD domains than the root zone? AD is more of an SMB thing than an NFS thing, but AIUI each zone can be its ow

Re: [discuss] ip6 && ip4 address listening inconsitency

2019-12-17 Thread Dan McDonald
Thank you! Alexander --> back to your question: > I understand correctly that the only way to avoid getting such duplicate mib > entries for ip4/ip6 sockets from /dev/arp via T_OPTMGMT_REQ is to eliminate > them on application level? I think the answer is "yes". :-/ Dan --

Re: [discuss] ip6 && ip4 address listening inconsitency

2019-12-17 Thread Dan McDonald
> On Dec 17, 2019, at 12:29 PM, Alexander Pyhalov wrote: > > Hi. > What do you have in ListenAddress ? If "::", it's expected. Looks like it's commented out. # grep Listen /etc/ssh/sshd_config #ListenAddress 0.0.0.0 #ListenAddress :: > Look at rpcbind, do you see 3 listen entries for it? >

Re: [discuss] ip6 && ip4 address listening inconsitency

2019-12-17 Thread Dan McDonald
> On Dec 17, 2019, at 2:09 AM, Alexander Pyhalov via illumos-discuss > wrote: > > We see two IPv4 listen entries and one IPv6 listen entry. > On Linux this porgram fails on second bind() with 'Address in use'. > > I don't know if it's a problem, likely, just a peculiarity. A lot of programs

Re: [discuss] [developer] Code of Conduct updates

2019-10-22 Thread Dan McDonald
Cutting to the crux here, if I missed something that wasn't clarification, I'm sorry. > On Oct 22, 2019, at 4:16 PM, caleb st. john wrote: > > > When you said: > > We will not be striking this term, as doing so would hide the problem that > clearly exists. > > This was confusing, to me, bec

Re: [discuss] [developer] Code of Conduct updates

2019-10-22 Thread Dan McDonald
Two notes: Before I proceed, I know openzfs is in this thread, but I'm only speaking pertaining to illumos. I am in meetings all week, and clearly the shortness-of-cycles has and will show here. I'm addressing 3 mails in one response. Let's start with Udo: > I fully understand that this is s

Re: [discuss] [developer] Code of Conduct updates

2019-10-22 Thread Dan McDonald
> On Oct 22, 2019, at 11:50 AM, Udo Grabowski (IMK) > wrote: > > "Race" is a term that itself already stems from at least some form of > racism. There are no "races" in the homo sapiens sapiens subspecies, to > which all living humans belong. Especially as a German, I'm always bewildered > to

Re: [discuss] Illumos and SegmentSmack/FragmentSmack?

2018-08-15 Thread Dan McDonald
Sending to developer too... > On Aug 15, 2018, at 9:41 AM, Bob Friesenhahn > wrote: > > Is the Illumos IP stack known to suffer from IP and/or TCP segment reassembly > DOS problems as FreeBSD and Linux have recently been found to have? I had early access to a PoC, but could not make it cause

[discuss] HEADS UP: Removal of 32-bit x86 kernel

2018-01-25 Thread Dan McDonald
Robert wrote the heads-up, I'm just sending it. -- Dan McD. From: Robert Mustacchi Heads Up: Removal of 32-bit x86 kernel As previously announced and discussed on the developer mailing list and with distributions, the x86 32-bit kernel is being removed. After pushing '8685 uts an

Re: [discuss] new ucode incompatibility ?

2018-01-11 Thread Dan McDonald
> On Jan 11, 2018, at 11:32 AM, Dan McDonald wrote: > > > >> On Jan 11, 2018, at 11:06 AM, Udo Grabowski (IMK) >> wrote: >> >> Hello, >> >> regarding commit 1adc3f: 8958 Update Intel ucode to 20180108 release, >> >> I tried to

Re: [discuss] new ucode incompatibility ?

2018-01-11 Thread Dan McDonald
> On Jan 11, 2018, at 11:06 AM, Udo Grabowski (IMK) > wrote: > > Hello, > > regarding commit 1adc3f: 8958 Update Intel ucode to 20180108 release, > > I tried to load the new intel-ucode.txt file with ucodeadm (oi151a9 and > hipster illumos-a90d75b), it just spits 'File header is invalid', >

Re: [discuss] A modern laptop for Illumos development?

2017-10-31 Thread Dan McDonald
If you can translate Linux to illumos, here: https://wiki.archlinux.org/index.php/Laptop/Asus has a breakdown of Linux-happy drivers. r8169 is a realtek chip, so I'm guessing rge(7D). rge sources indicate some r8169 support, so at least there you're good. There's a warning about Intel

Re: [discuss] Where do we need innovation?

2017-09-14 Thread Dan McDonald
> On Sep 14, 2017, at 10:14 PM, Alex Smith (K4RNT) > wrote: > > Could the latest LX branded Zones from SmartOS be ported to the rest of > Illumos, or other distributions such as NexentaStor or OpenIndiana? I'd like > to have Linux branded Zones in OpenIndiana. We did it with OmniOS. Dan --

Re: [discuss] vbox vs kvm

2017-07-20 Thread Dan McDonald
> On Jul 20, 2017, at 7:26 AM, Jim Klimov wrote: > > VBox VMs as a process can run in zones, benefitting from more resource > controls and access separation, admin delegation etc. if needed. Note also my > vboxsmf project to run VMs as SMF instances. So can KVM. Dan

[discuss] OCR software for illumos distros?

2017-04-24 Thread Dan McDonald
Lets say I scan in a bunch of paper at 300dpi or whatever. I'd like to turn it into ASCII. Does such an OCR beast exists that's portable to, or already running on, illumos? Also, something that works in an LX zone is acceptable if there's nothing native. Thanks, Dan --

[discuss] Fwd: VMware Fusion 8.5.6 + macOS 10.12.4 == black screens

2017-03-29 Thread Dan McDonald
I sent this to the OmniOS discussion list, but someone reminded me this affects ALL illumos distros. Please read below. Thanks, Dan > Begin forwarded message: > > From: Dan McDonald > Subject: VMware Fusion 8.5.6 + macOS 10.12.4 == black screens > Date: March 29, 2017 at 11:30

Re: [discuss] Pkg Versions

2017-03-21 Thread Dan McDonald
Hello! If you have OmniOS-specific questions, you should join the OmniOS discussion list: http://lists.omniti.com/mailman/listinfo/omnios-discuss But since you're here, and my answers also apply to OpenIndiana... > I am just being introduced to IllumOS and OmniOS. I am coming from Free

Re: [discuss] Latest master build fail due to getxby_door.h missing

2017-02-24 Thread Dan McDonald
> On Feb 24, 2017, at 4:42 PM, Saso Kiselkov wrote: > > I'm pretty sure I didn't prepend anything to it, though, is chances are, > the default install will fail the build as well. r151014(usr/src)[1]% sudo cat /root/.profile # # Simple profile places /usr/gnu/bin at front, # adds /usr/X11/bin,

Re: [discuss] Latest master build fail due to getxby_door.h missing

2017-02-24 Thread Dan McDonald
> On Feb 24, 2017, at 4:30 PM, Saso Kiselkov wrote: > > > ../port/gen/getgrnam_r.c:59:25: error: getxby_door.h: No such file or > directory > *** Error code 1 > > There's no -I../port/gen which contains the required header file. I've > not yet traced my way all through the makefiles, so no ide

Re: [discuss] Latest master build fail due to getxby_door.h missing

2017-02-24 Thread Dan McDonald
> On Feb 24, 2017, at 4:12 PM, Dan McDonald wrote: > > Interesting you mention OmniOS. I built illumos-gate just today on OmniOS > bloody, but I'll spin up builds on both my build boxes for releases r151014 > and r151020 to see if anything stands out. I will point ou

Re: [discuss] Latest master build fail due to getxby_door.h missing

2017-02-24 Thread Dan McDonald
> On Feb 24, 2017, at 4:07 PM, Saso Kiselkov wrote: > > place. Any ideas what I'm doing wrong? Interesting you mention OmniOS. I built illumos-gate just today on OmniOS bloody, but I'll spin up builds on both my build boxes for releases r151014 and r151020 to see if anything stands out. Dan

Re: [discuss] ZFS recv hangs

2016-11-15 Thread Dan McDonald
As discussed on the OpenZFS slack, I've seen this bug affect OmniOS r151014 and r151016. Now note that r151014 has some backported bugfixes in it from later releases, BUT I know that r151018 does not have this problem. The send-stream I have which tickles this bug is the current OmniOS bloody Z

Re: [discuss] licensings

2016-11-08 Thread Dan McDonald
>> Hi, anyone has good knoweledge anout open source licensings? > On Nov 8, 2016, at 11:35 AM, Joerg Schilling > wrote: > > - The CDDL ist the best reciprocal (copy left) license I know Joyent has been recently putting out non-illumos stuff under Mozilla Public License 2.0. CDDL was a de

[discuss] OmniOS r151020 is now out!

2016-11-04 Thread Dan McDonald
ding (but not limited to): Dale Ghent, Marissa Murphy, and Robert Treat (who convinced me to port LX). Thanks to Joyent staffers Jerry Jelinek and Patrick Mooney, whose LX work AND LX-porting-assitance were invaluable. Happy upgrading! Dan McDonald -- OmniOS Engineering -

Re: [discuss] i219-LM

2016-10-11 Thread Dan McDonald
> On Oct 11, 2016, at 10:59 AM, Eric Sproul wrote: > > Thanks Robert, I'm sorry for muddying the water with my misunderstanding. Likewise. Dan --- illumos-discuss Archives: https://www.listbox.com/member/archive/182180/=now RSS Feed: https://www.listb

Re: [discuss] i219-LM

2016-10-11 Thread Dan McDonald
> On Oct 11, 2016, at 9:54 AM, Eric Sproul wrote: > > I would try adding the PCI ID as "pci1849,15b7", as your motherboard vendor > seems to have used their own prefix, and maybe it's not reporting its > compatibility IDs properly. If that doesn't work, you may need to wait for > 7458 to int

Re: [discuss] default system shell

2016-08-29 Thread Dan McDonald
men! bloody(~)[0]% grep danmcd /etc/passwd danmcd:x::10:Dan McDonald:/export/home/danmcd:/usr/bin/tcsh bloody(~)[0]% Dan --- illumos-discuss Archives: https://www.listbox.com/member/archive/182180/=now RSS Feed: https://www.listbox.com/member/archive/rss/

[discuss] Re: LX on OmniOS update

2016-08-26 Thread Dan McDonald
I've updated all of the following URLs: > On Jul 1, 2016, at 3:04 PM, Dan McDonald wrote: > First some URLs. The git repos are: > > https://github.com/danmcd/io-lx-public/ > > https://github.com/danmcd/ob-lx-public/ > > The latter has only one small

Re: [discuss] Re: [developer] Re: Request: Formal permanent lifetime ban on Martin Bochnig (*not* Martin Sustrik)

2016-08-18 Thread Dan McDonald
> On Aug 18, 2016, at 10:30 AM, Joerg Schilling > wrote: > > There is more than one single person who is responsible for the problems. We may suffer from an exclusivity because of who's ex-Sun vs. who's not-ex-Sun - I won't deny that. BUT I don't see how that has anything to do with honest-

Re: [discuss] Removing gendered pronouns from illumos source

2016-08-01 Thread Dan McDonald
> On Aug 1, 2016, at 4:36 PM, Dj Hoffman wrote: > > Hello all, > At Delphix we are working on a patch to remove gendered pronouns > (he/she/his/hers/him/her/himself/herself) in favor of gender-neutral singular > pronouns (they(1)/them/themself(2)/it/its) when they don't reference specific > p

[discuss] LX on OmniOS update

2016-07-01 Thread Dan McDonald
Hello! I'm BCc:ing the illumos developer and illumos discussion lists. This thread should continue on omnios-discuss. Please note I will be very slow to get back to emails until July 12th. I'd like to report some progress on the LX zones for OmniOS work. First some URLs. The git repos are:

Re: Solved: [discuss] krtld failed to open, again...

2016-06-15 Thread Dan McDonald
> On Jun 15, 2016, at 2:30 AM, Gabriele Bulfon wrote: > > This is very new to me. How can we let the system warn that the menu.lst is > getting too big, potentially breaking boot? > Short-term: beadm list -H| wc -l and report a number over a threshhold. We don't enforce or warn in our to

[discuss] FLAG DAY - 4719 affects nightly, package, and poold

2015-05-05 Thread Dan McDonald
Illumos #4719 introduces a flag day for people who build illumos-gate. Starting now, you will need a Java Developers Kit (JDK) 7 or later. OpenIndiana 151a9 does NOT have this by default. Builders must either set JAVA_ROOT to a source of JDK7, or must have /usr/java populated with JDK7. Users sti

[discuss] HEADS UP -- illumos-gate can now be built on OmniOS r151014 or later

2015-05-05 Thread Dan McDonald
With the pushes of 4719, 5878, and 5879, one may now build stock illumos-gate on OmniOS, revisions r151014 or later. An OmniOS .env file will need certain variables set. I'm attaching a sample one I use, but I will go over the critical variables here. To build on OmniOS, you must: 1.) Use "gcc

Re: [discuss] Huge performance penalty with aggr over ixgbe

2014-06-24 Thread Dan McDonald via illumos-discuss
On Jun 24, 2014, at 8:59 AM, Richard Elling via illumos-discuss wrote: > Kris, > are deep C-states enabled? If they are, disable them (via the BIOS). You mention a CPU difference between your two boxes. Are they the same socket-type and chipset as well? Dan ---

Re: [discuss] Fwd: MPT SGL mem alloc failed

2014-05-28 Thread Dan McDonald via illumos-discuss
On May 28, 2014, at 1:22 PM, Garrett D'Amore via illumos-discuss wrote: > I suspect the root issue was never resolved. You should review your memory > allocation; also, older mpt adapters may have a requirement to use memory for > DMA with certain restrictions. This memory can become scarce

Re: [discuss] panic assertion failed: !IN6_IS_ADDR_V4MAPPED(dst_addr), file: ../../common/inet/ip/ip6.c, line: 1984

2014-02-13 Thread Dan McDonald
On Feb 10, 2014, at 1:17 PM, Richard PALO wrote: > This is the first time I've come across this: >> panic[cpu1]/thread=ff02e79e7880: >> assertion failed: !IN6_IS_ADDR_V4MAPPED(dst_addr), file: >> ../../common/inet/ip/ip6.c, line: 1984 Richard provided me a coredump, and I've filed bug 45

Re: [discuss] panic assertion failed: !IN6_IS_ADDR_V4MAPPED(dst_addr), file: ../../common/inet/ip/ip6.c, line: 1984

2014-02-10 Thread Dan McDonald
On Feb 10, 2014, at 1:17 PM, Richard PALO wrote: > This is the first time I've come across this: >> panic[cpu1]/thread=ff02e79e7880: >> assertion failed: !IN6_IS_ADDR_V4MAPPED(dst_addr), file: >> ../../common/inet/ip/ip6.c, line: 1984 >> >> >> ff0010671340 genunix:process_type+162480

[discuss] PTPd for Illumos?

2014-01-14 Thread Dan McDonald
Consider the Precision Time Protocol (PTP), especially as implemented here: http://ptpd.sourceforge.net/ Has anyone spent any cycles getting this up and running on Illumos? I have a customer inquiring about it, but I don't want to duplicate any effort. Thanks, Dan --

Re: [discuss] SMB 2.1 support in Illumos

2013-12-19 Thread Dan McDonald
On Thu, Dec 19, 2013 at 07:58:14AM -0800, Garrett D'Amore wrote: > Generally speaking, if they have modified CDDL files in the course of > adding support, then they are obliged to make their changes to those files > available to their customers (which in the case of a free download can be > *anyone

Re: [discuss] Highpoint Non-Raid r750

2013-10-08 Thread Dan McDonald
Dumb question: I may have missed this earlier, but are you controlling large numbers of SAS drives? Or SATA drives? If "SATA drives", I'd be very wary. There are others on this list who can speak to the dangers of using a SCSI HBA to control SATA drives. Dan ---

Re: [discuss] Highpoint Non-Raid r750

2013-10-08 Thread Dan McDonald
On Tue, Oct 08, 2013 at 04:05:29PM +0100, Jonathan Adams wrote: > would 2 of the 9201 16i cards : > http://www.scan.co.uk/products/16-port-lsi-sas-9201-16i-sgl-6gb-s-sata-plus-sas-pcie-20-controller > > work, in your opinion? This *SHOULD* work, especially if it has the "IT firmware" on it. Dan

Re: [discuss] Highpoint Non-Raid r750

2013-10-08 Thread Dan McDonald
On Tue, Oct 08, 2013 at 03:31:03PM +0100, Jonathan Adams wrote: > I've found the LSI MegaRAID 9280 24i4e (since we currently have 6 backplane > boards, of 4 drives each, this would probably do) > > about ?1000, so only 20% more ... but do you know if such a board will do > JBOD? Such a board will

Re: [discuss] Firewire Audio

2013-09-11 Thread Dan McDonald
On Wed, Sep 11, 2013 at 04:33:04PM +0200, Gabriele Bulfon wrote: > I just connected the firewire to the machine, turned on the mixer. Nothing. > Is there any tool in illumos so I can scan the system for firewire attached > devices, low level? What does "cfgadm -l" say? Also please look at the ie

Re: [discuss] Supermicro X9 series Mobo with SAS and 10GB ethernet

2013-05-22 Thread Dan McDonald
On May 22, 2013, at 3:40 PM, Geoff Nordli wrote: > > It looks like Chris has flashed it, but it is probably better to stay away > from the 2208. > > What form factor did you find the X540 and 2308? It doesn't even need to be > a dual socket machine. This guy: http://www.supermicro.

Re: [discuss] Supermicro X9 series Mobo with SAS and 10GB ethernet

2013-05-22 Thread Dan McDonald
On May 22, 2013, at 2:11 PM, Geoff Nordli wrote: > Anyone running one of these: > > http://www.supermicro.com/products/motherboard/xeon/c600/x9drh-7tf.cfm > > From seeing the discussion a few months back Saso was about to buy one. > > Any links to the instructions on reflashing the LSI 2208 to

Re: [discuss] Fw: [solarisx86] ZFS and encryption

2013-03-26 Thread Dan McDonald
On Tue, Mar 26, 2013 at 05:13:27PM +0100, Jesus Cea wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > On 26/03/13 16:52, Nick Zivkovic wrote: > > Per dataset would be ideal. But full disk is better than nothing. > > I fully agree here. > > But... I found annoying that Oracle ZFS encry

Re: [discuss] Fw: [solarisx86] ZFS and encryption

2013-03-26 Thread Dan McDonald
Is there no interest in per-dataset keying? Are people too afraid of butting heads with Oracle? Is it Too Hard (TM)? I've not seen enough paying-customer interest, though it's possible such interest isn't reaching me. I see peaks on public lists like now, but not with the fervor to justify a

Re: [discuss] Performance on new quad socket system

2013-03-25 Thread Dan McDonald
On Mon, Mar 25, 2013 at 09:43:31PM +0100, Sa?o Kiselkov wrote: > On 03/25/2013 06:39 PM, Franz Schober wrote: > > Hi, > > A bunch more things came to my mind, see comments below: > > > Test: Writing a 1 GB File > > time dd if=/dev/zero of=/tmp/testfile1 bs=128k count=8k > > Mean value of 3 tests

Re: [discuss] Re: [OpenIndiana-discuss] hardware specs for illumos storage/virtualization server

2012-11-17 Thread Dan McDonald
On Nov 17, 2012, at 11:17 AM, Bob Friesenhahn wrote: > I have a OpenIndiana system with 16-cores of Xeon E5 and Intel i350 Ethernet > and these seem to work great with recent OpenIndiana/Illumos. I was a bit > nervous about i350 Ethernet since the driver was not added until 151a3 but it > has

Re: [discuss] Kazuyoshi tun0, zones, blowfish

2012-10-15 Thread Dan McDonald
On Mon, Oct 15, 2012 at 10:28:15AM +0200, Gabriele Bulfon wrote: > > I tried adding the same /dev/tun to 2 different zones, and it didn't > complain, both zones boot. One I start openvpn on one zone, I can't start > it on the other zone (complaining that it can't create the interface > because "f

Re: [discuss] Home server conundrum follow-up

2012-09-17 Thread Dan McDonald
On Mon, Sep 17, 2012 at 08:15:12PM +0200, Jesus Cea wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > On 17/09/12 20:08, Andrew Haycock wrote: > > Well, I lie, there are a couple of questions but I don't think > > they're too taxing. My concern with the ZIL being located on the > > SSD

Re: [discuss] Home server conundrum follow-up

2012-09-17 Thread Dan McDonald
On Mon, Sep 17, 2012 at 07:08:27PM +0100, Andrew Haycock wrote: > Dan, that helped a lot more at least to the point that any further > questions are quite reasonably discoverable with experimentation or at > least google fodder. Cool. I like happy (potential) users. > Well, I lie, there are a co

Re: [discuss] Home server conundrum follow-up

2012-09-17 Thread Dan McDonald
Ooops, I needed to clarify something... On Mon, Sep 17, 2012 at 06:26:12PM +0100, Andrew Haycock wrote: > When Dan explained his setup, he brought to my attention his 64GB root > disk. Aha! I think, this is something I was considering, but then I > come away thoroughly confused as I try to unde

Re: [discuss] Home server conundrum

2012-09-17 Thread Dan McDonald
I've been using Solarish platforms for my home server for the past four years on the same piece of 2006-era HW. Start by looking here: http://kebesays.blogspot.com/2008/03/kebes-home-data-center-or-fbarts-new.html And realize now that I've upgraded/changed things as follows: - OS was in-develop

Re: [discuss] Driver location question

2012-08-30 Thread Dan McDonald
On Thu, Aug 30, 2012 at 12:04:48PM -0700, Michael Speer wrote: > All, > > Are there any differences with having a file reside in /usr/kernel > versus having > them reside /kernel? What is the difference between the two locations and > why would one location be chosen over the other? In the REAL

Re: [discuss] Problem with multicast fanout and Rx queues

2012-06-28 Thread Dan McDonald
I thought 918 wouldn't be the issue here as there is no VLAN involved. ANYWAY, I do have a just revised today 918 and its webrev (because one of our Nexenta customers was bitten by 918) here: http://kebe.com/~danmcd/webrevs/918/ You should be able to drop these diffs into your illumos-g

Re: [discuss] Problem with multicast fanout and Rx queues

2012-06-27 Thread Dan McDonald
On Jun 27, 2012, at 8:23 AM, Sašo Kiselkov wrote: > On 06/27/2012 01:47 PM, Udo Grabowski (IMK) wrote: >> ip_squeue_fanout=1 ? > > Thanks for the tip. Just tried it even with a reboot and it didn't help. > My cpu23 is still being hammered hard. Back in S10 there used to be a > ip_soft_rings_cnt

Re: [discuss] XHCI driver

2012-06-16 Thread Dan McDonald
On Jun 16, 2012, at 12:37 PM, Apostolos Syropoulos wrote: > > > Hello, > > I have seen the following: > > http://svnweb.freebsd.org/base?view=revision&revision=213379 > > which is an implementation of the XHCI driver for FreeBSD. > Is it possible to port this? I mean is it easy or difficult,

Re: [discuss] Illumos on Supermicro X9DAi motherboard

2012-06-15 Thread Dan McDonald
On Fri, Jun 15, 2012 at 09:10:26AM -0500, Bob Friesenhahn wrote: > This is what my computer builder tells me is still not supported by > development OpenIndiana on this motherboard: > > USB 3.0 chip Correct. USB3 is going to be non-trivial. > C600 chipset thermal management > Last 4 ports

Re: [discuss] Is it possible to do shared networking for LZ different from IP stack in GZ?

2012-06-11 Thread Dan McDonald
The fundamental question is always: What problem are you really trying to solve? On Jun 11, 2012, at 9:59 AM, Jim Klimov wrote: > Hello all, > > In OpenSolaris and its descendants it is possible to create > local zones (LZ) which share an IP stack with the global zone > (GZ) or have an exclus

Re: [discuss] User-space locks slow on Opteron 6200?

2012-04-11 Thread Dan McDonald
On Wed, Apr 11, 2012 at 12:35:09PM -0500, Bob Friesenhahn wrote: > Your suggestion of 'plockstat' was excellent. Different application > algorithms show different high-runners but the high-runner locks are > usually not called very often but are held for an abnormally long > time. For some alg

Re: [discuss] Illumos on Xeon E5-2600 platform?

2012-03-15 Thread Dan McDonald
On Thu, Mar 15, 2012 at 09:36:08AM -0500, Bob Friesenhahn wrote: > Is Illumos prepared to boot and run properly on Intel's new > two-socket Xeon E5-2600 configurations? This is what's known as Romley, right? > Does Illumos provide drivers which will work with the SATA ports offered by > the Intel

Re: [discuss] Re: [OpenIndiana-discuss] Illumos does not boot on Intel core 5i desktops

2012-02-21 Thread Dan McDonald
recent OI update does contain some acpica changes that might help: changeset: 13539:15b6c91053a3 user:Gordon Ross date:Thu Dec 01 22:58:02 2011 -0500 description: 1621 update acpica code to newer version (Joyent OS-441) Reviewed by: Garrett D'Amore

Re: [discuss] DEBUG enabled

2012-02-20 Thread Dan McDonald
On Mon, Feb 20, 2012 at 03:44:41PM +0100, Gabriele Bulfon wrote: > Hi, > following suggestions from the list to disable the debug kernel, I changed my > illumos.sh > NIGHTLY_OPTIONS removing D and F: > export NIGHTLY_OPTIONS='-nCmprt' > then ran a full rebuild, and onu install. > On rebooting insi

Re: [discuss] shutting down the forums on illumos.org

2012-02-09 Thread Dan McDonald
On Thu, Feb 09, 2012 at 09:50:55AM -0800, Deirdre Straughan wrote: > Ideally, we would also keep an eye on the list for newbie FAQs, and answer > them on the wiki in a systematic fashion. I hope that the new wiki will > make it a little easier for community members to contribute documentation > of

Re: [discuss] shutting down the forums on illumos.org

2012-02-08 Thread Dan McDonald
On Wed, Feb 08, 2012 at 04:07:26PM -0800, Deirdre Straughan wrote: > Any objections to shutting these down? None whatsoever. Nuke 'em. Dan --- illumos-discuss Archives: https://www.listbox.com/member/archive/182180/=now RSS Feed: https://www.listbox.com

Re: [discuss] PLEASE REVIEW: Matching light/dark Illumos web logos

2012-01-28 Thread Dan McDonald
Okay... I hope we can avoid bikeshedding. I like the Marsell-improved logos, though I do miss the "powered by" in the 32px version (even if it is tiny... I can read it). Anyway, here are the two URLs (in order of best-looking to least-best-looking): http://desacci.org/illumos-logos/

Re: [discuss] PLEASE REVIEW: Matching light/dark Illumos web logos

2012-01-28 Thread Dan McDonald
On Sun, Jan 29, 2012 at 11:46:34AM +1100, Marsell Kukuljevic wrote: > > I'll gladly take better versions. > > http://desacci.org/illumos-logos/ These are light-background only. Can you do the dark-background one as well? > It's a pity the illumos logo doesn't scale down well. These used to be >

Re: [discuss] PLEASE REVIEW: Matching light/dark Illumos web logos

2012-01-28 Thread Dan McDonald
On Sun, Jan 29, 2012 at 09:25:20AM +1100, Marsell Kukuljevic wrote: > > 1.) Feedback. Does it look okay? Would you like to see a website sporting > >one of these? > > The smallest one is hard to read. > > I'd suggest using Lanczos resampling and an unsharp mask on both resizes, > and removi

  1   2   >