Activity
From 2012-02-27 to 2012-03-27
2012-03-27
-
11:20 PM Bug #2248: assertion failed: ilu->ilu_ntasks_min_free <= ilu->ilu_ntasks_free
- It crashed again on Sunday night. 4G is still not enough, and there's not enough room on the VMDK to give it more sp...
-
11:14 PM Bug #2194 (Resolved): ses module for libfm does not ignore return values correctly
- Resolved in r13648 commit:5d8678c9e68e
-
06:20 PM Bug #2194 (Pending RTI): ses module for libfm does not ignore return values correctly
-
- Resolved in r13649 commit:049724836164
-
07:22 PM Bug #2449 (Feedback): Add workaround for AMD K10 CPU erratum 721
- Webrev: http://cr.illumos.org/view/2m34dd4u/illumos-2449-webrev/
-
06:58 PM Bug #2536: iscsi: idm unexpectedly closes socket connection due to EFAULT in uiomove
- Sorry, I posted this bug in the wrong bug tracker...
-
- OS: oi_151a
when issuing SCSI read commands to a iscsi target on Linux (SCST 2.1.0, scst_cdrom module), I regularl... -
- I added WiFi extender to my home but now I am impacted by bad algorithm, selecting AP "randomly", ignoring the one wi...
-
06:31 PM Bug #2402: rnode.r_lkserlock is stupid (NFS4)
- Would be good to figure out why the NFS code is holding this lock here, while calling out for a remote lock. Usually...
-
06:23 PM Feature #650 (In Progress): grep support for -q would be useful
-
03:40 PM Bug #2547 (Resolved): Minor updates to README
- The top-level README needs a few updates. Probably the whole thing should be rewritten, but for now we can just fix ...
-
- Fixed in:
changeset: 13647:395e51435a78
tag: tip
user: enricop <enricop@computer.org>
date:... -
12:45 PM Bug #2401: buffer free to wrong cache
- Another data point. I also have this issue with my IBM M1015 (SAS2008) controller with the default (IR) firmware usi...
-
- Webrev: http://cr.illumos.org/view/ytr0djc9/illumos-2546-webrev/
- updated AMD microcode patch file (webrev seems ... -
- The format of the AMD microcode patches has changed for family 0x14 and 0x15, patches for those families are now encr...
-
07:03 AM Bug #2544: mpt_sas substract long from int32_t (64Bit)
- I mean it is wrong from a theoretical perspective :-)
-
- I don't have enough experience in order to know what kind of values mptsas_scsi_watchdog_tick will take in practice. ...
-
- But only if mptsas_scsi_watchdog_tick has a value which doesn't fit in 32bits, which would be a really nonsensical va...
-
- It was found by statical analysis. But practical problems will arise when timeouts are not getting handled.
-
- Are you seeing practical problems stemming from this, or was this found by static analysis?
-
- Several occurences where numeric datatypes of different size are substracted from each other
for example:
int32... -
- Duplicate bug id, closed are request of submitter.
2012-03-26
-
03:14 PM Bug #2543 (New): bad duality in metaslab_df_alloc/metaslab_df_fragmented
- in metaslab_df_alloc the rule to switch to first-fit is:
if (max_size < metaslab_df_alloc_threshold ||
... -
- activate zfs_prefetch_disable disable all dmu_prefetch.
For metaslab_prefetch is a very bad idea:
When zfs decide... -
05:30 AM Bug #2538 (New): ipfilter breaks fragment cache table
- Ipf fragment cache table is protected by rwlock.
Although fr_fraglookup() modifies the cache list, it is called with... -
- Sorry, I updated work around patch....
-
- How to reproduce.
1. On nfs server, allow from nfs clients... -
01:30 AM Bug #2401: buffer free to wrong cache
- Albert Lee wrote:
> This is a known issue in imr_sas, which uses a struct scsi_inquiry size from Solaris 10. Nexenta... -
- OK, I will update the driver and have a check.
2012-03-25
-
- OS: oi_151a
when issuing read commands to a iscsi target on Linux (SCST 2.1.0, scst_cdrom module), I regularly see... -
- For years sendmail, postfix and other smtp servers if they want to access nsswitch aliases database, they have to use...
-
- Not only that. Upstream abandoned HAL, so all its development is on us now anyway. Some of Solaris specific parts wer...
-
- Because all the important parts had to be written fresh for the Solaris port, and it (in theory, I haven't checked re...
-
- Resolved in r13645 commit:1cef92b663d9
-
- Resolved in r13646 commit:8f5bd7d387bc
-
- Resolved in r13646 commit:8f5bd7d387bc
-
03:52 AM Bug #2401: buffer free to wrong cache
- This is a known issue in imr_sas, which uses a struct scsi_inquiry size from Solaris 10. Nexenta has reported this is...
-
01:28 AM Bug #2430: panic: BAD TRAP under heavy load with mpt_sas
- I mean, I can host the crashdump easily, I just wanted to explain why I didn't initially post it.
http://skysrv.ph...
2012-03-24
-
10:09 PM Bug #2127: 'zpool create' core dumps
- http://dl.dropbox.com/u/7506135/illumos%20%232127/core_zpool.bz2
-
- Yes I have.
-
- Do you have the core file available?
-
09:38 PM Bug #2071: "libhal_device_free_changeset" function mem-leeks.
- Why HAL is maintained in illumos-gate? It should be in userland.
-
- It seems there are few leaks more in the code, I will look at them.
-
07:23 PM Feature #2451 (Resolved): beadm umount should report if there is nothing to unmount
- Currently, "beadm umount" reports about successful unmounting of the ABE, even if it was already unmounted. While thi...
-
- Currently, "beadm umount" only supports the BE name as an argument. I see little reason why it can't also accept the ...
-
- The size of crashdump is really big - did you enable taking userspace pages also?
If I provide some space for uplo... -
- ...
-
- A single MSR write resolves erratum #721 described here:
http://support.amd.com/us/Processor_TechDocs/41322_10h_Rev_...
2012-03-23
-
- Several occurences where numeric datatypes of different size are substracted from each other
for example:
int32... -
- be_mount just return "BE_ERR_INVAL" for these cases (but should different), which not processed in be_do_mount and ...
-
- beadm provides extended error diagnostics with a preset environment value "BE_PRINT_ERR=true", but that does not see...
-
- # savecore -vf /var/crash/myshost/vmdump.3
savecore: bad magic number 3a650c00
# echo $?
0
And looking through ... -
08:45 AM Bug #405 (In Progress): Can't connect to similarly-named, encrypted wireless networks.
- the nwam behaviour on 151.1.2 is the same
after setting up and enabling a new ncp with a configured ncu phys/ip bo... -
- It's not clear to me what triggered it - this machine has never been below oi_151, but it also hasn't had this recur ...
-
12:18 AM Bug #2430: panic: BAD TRAP under heavy load with mpt_sas
- Hi Rich,
Can you provide a bit more information - is this a fresh install or has this machine been running fine un...
2012-03-22
-
11:30 PM Feature #862 (Feedback): KMS support in OI
- This relates to https://www.illumos.org/issues/577 and https://www.illumos.org/issues/1524 for radeon users.
The ATI... -
08:10 PM Bug #2443 (Resolved): wall(1) needs zones support
- wall(1) only broadcasts to the global zone. It would be useful to send messages to the ngz.
-
2012-03-21
-
- the manifest patch is attached
-
- This model is a LinkSYS PCI card with a RaLink RT2860 chip
With the stock rwn driver in oi-151a2 prestable1 the nic ... -
01:35 PM Feature #2433 (Resolved): mount: allow relative path
- /sbin/mount utility doesn't apply relative path as a mount point. That's very annoying.
Example:... -
05:37 AM Bug #2408: CJK character width handled incorrectly in terminal emulators
- My locale is ja_JP.UTF-8, but the problem should also manifest in English locales. As for sample text, try:
これは例です... -
- Yeah, so looking at this more, we missed it. So did the POSIX committee apparently -- there is no standard way to pr...
-
- It would help me if you could provide specific detail about what locale you were using, and sample Japanese text to w...
2012-03-20
-
- Resolved in:
changeset: 13642:8107b374f945
tag: tip
user: Vitaliy Gusev <gusev.vitaliy@nexent... -
- uname -a:...
2012-03-19
-
- I've set the size of the dump volume to 4G to see if I can get a complete dump. mdb -k shows I'm currently using som...
-
- No, I didn't make any changes.
The only interesting thing in messages:
@Mar 19 14:36:59 thor drmach_acpi: [ID 779... -
08:23 PM Bug #2248: assertion failed: ilu->ilu_ntasks_min_free <= ilu->ilu_ntasks_free
- Will Murnane wrote:
> This problem has recurred several times. I have uploaded another core file to the same locati... -
- Will Murnane wrote:
> I should mention I'm running a custom-built kernel. Here's the tree I built it from:
> $ hg ... -
- This problem has recurred several times. I have uploaded another core file to the same location as before; look at t...
-
09:05 PM Feature #1469: ttyc/ttyd should be an allowed console device
- An updated webrev prepared by Gary Mills on 2012-Mar-10 located at It still works for me in a 2012-03-19 nightly buil...
-
- The 64bit cpr wakecode works around bugs in the Studio 10 (Vulcan) assembler, related to the generation of 16bit code...
-
01:47 AM Bug #2401: buffer free to wrong cache
- eric zhang wrote:
> the driver from LSI, version. 3.03_2008_Solaris
>
> from the screenshot, we saw it should be ...
2012-03-18
-
11:54 PM Bug #2426: wanboot openssl needs to not call cast function pointers
- can we use static libssl.a & libcrypto.a from userland build ?
I have temporary fixed illumos-build on sparc for lib... -
- Casting of function pointers is undefined.
GCC feel strongly about this, apparently, to the degree that it 4.2 the... -
- We don't ever use Sun as on amd64, but we have code to support it specifically some of which is screwy, and which unl...
-
- The problem on the 32bit side is that by sheer bad luck the Sun as negation syntax is such that GNU as parses it as c...
-
- Resolved in r13640 commit:2cb107f54874
-
- Resolved in r13641 commit:9efb7f317b8a
2012-03-17
-
- Macro replaced by function.
-
11:08 AM Bug #1765: assert triggered in libzfs_import.c trying to import pool name beginning with a number
- The original title was misleading because the problem has nothing to do with regex.
-
06:39 AM Bug #1765: assert triggered in libzfs_import.c trying to import pool name beginning with a number
- Original title was correct:...
2012-03-16
-
- This is fun.
On x64 we always use gas, via the assembler wrapper, 'aw'. Because of Randy's typo, this means that ... -
- Sod it, should be fun
-
- I may have this dead wrong, but want to leave it until I'm sure.
-
- When debugging something else, I've noticed that on Illumos, and as best as I can tell only Illumos, the "dumb" termi...
-
- A combination of crappy circumstance left me with accidentally debug builds. We only 'install -s' when -nd
-
- Looking at recently builds, on both platforms, I've noticed that we're getting certain .debug_* sections in certain o...
-
- Bart's answer is pertinent I think: http://thread.gmane.org/gmane.os.solaris.opensolaris.networking/1318/focus=1326
... -
08:21 AM Bug #2414: IOV_MAX = 16?? low?
- question has been around:
http://article.gmane.org/gmane.os.solaris.opensolaris.networking/1318 -
- Came across this (http://lists.unbit.it/pipermail/uwsgi/2011-May/002082.html) and checked the sources:...
-
- On systems supporting AVX, we need to preserve the %ymm registers in elf_rtbndr as we currently do with %xmm, to prev...
-
- Resolved in r13638 commit:8dacb0cc3292
-
- Resolved in r13639 commit:4c26102b6754
-
- There are a number of CPU features which, in cpuid.c are conditional on Intel being the processor vendor. Several of...
-
- The bug submitter was apparently very confused about this.
- context switch support is done (that's xsave/xrstore)...
2012-03-15
-
- Such milestone is probably part of sysidcfg replacement and comes from the overall design of S11. I am not sure what ...
-
- in S11 Oracle added an SMF milestone which comes online when system configuration and self-assembly is complete (that...
-
05:52 PM Bug #2410 (Resolved): procfs needlessly breaks large file support
- When attempting to include @<sys/procfs.h>@ in compiling a 32-bit program with large file support, the compile will e...
-
- On 32bit sparc, libc needs to provide __floatundidf and __floatundisf as private interfaces, such that we can build w...
-
- Oh, one other thing: most modern devices have a fairly deep on-device queue, so we expect that if you have a reasonab...
-
- It applies to both reads and writes.
However, with typical access patterns, I suspect you might not see many conti... -
- Ummm... a sort of lame question: does disksort in "sd" only apply to reads, or to queued writes as well?
IIRC, the S... -
07:13 AM Bug #2399: zfs manual page does not document use of "zfs diff"
- Hi Jon, this is good investigative work here, but seems to be a duplicate of #545:
https://www.illumos.org/issues/545
2012-03-14
-
- On Illumos-based systems, CJK character width is not handled correctly in any terminal emulator. The terminal emulato...
2012-03-13
-
- Resolved in r13637 commit:739a180bd323
-
- I'd vastly prefer that, yes.
-
- Shouldn't it be splitted into two parts: mmap.s and mmap64.s? AFAIK both mmap and mmap64 should exists in libc.
-
- As best as I can tell, this code is just bad.
The largefile environment _when compiling libc_ shouldn't matter, ... -
- @usr/src/lib/libc/common/sys/mmap.s@
Snippet in question (mmap.s):... -
- The problem of kernel panicking with such double-entries in DDT (one of which is invalid) was further explored by me ...
-
- Quite possibly, that is it.
My original source-code checkout was done in January (2012-01-25), and the bug #2067 was... -
- Quite possibly, that is it.
My original source-code checkout was done in January (2012-01-25), and the bug #2067 was... -
- It is necessary to know head changeset you use in your private build. Can it be afftecd by https://www.illumos.org/is...
-
- Since recently, my troublesome home NAS is running a privately built illumos-gate as fetched from source in mid-Febru...
-
04:32 PM Bug #2233: zfs mount hangs system
- That's what I always get with kernel panics on oi_151a too.
Executing start method ("/lib/svc/method/svc-dumpadm")... -
03:21 PM Bug #2233 (Feedback): zfs mount hangs system
- Hardly. 1) the system gave "magic number 0" which is, I beleive, no dump found. 2) system is important, and is back i...
-
- Could you attach kernel crash dump to the issue?
-
- There's a debug printf in pcisch which causes compiler weirdness relating to implicit types
We do, in effect,
<p... -
08:50 AM Bug #2402 (New): rnode.r_lkserlock is stupid (NFS4)
- Lock requests to a file hosted on nfs4 (and probably also nfs3) can slow down or completely starve writes to that fil...
-
- Even if you run "savecore" command manually, no data are stored to /var/crash/ ?
-
- the driver from LSI, version. 3.03_2008_Solaris
from the screenshot, we saw it should be saved in /var/crash/vs2/,... -
- Which driver do you use for that MegaRAID? imr_sas from LSI?
From the screenshot you provided I see that savecore ... -
- MJ,
thanks.
kernel program not device driver, just one cache module.
SAS adapter: LSI MegaRAID SAS 9240-8i 2... -
- What is your "kernel program" (is it device driver?) doing?
Which SAS adapter do you use?
Is the crash reproduc... -
- sorry , my os infor.
uname -a
SunOS vs1 5.11 oi_151a2 i86pc i386 i86pc Solaris -
- We install the openindiana on raid0 SAS HDD, when we run our kernek program, the kernel crash, show "kernel memory a...
-
- There seems to be an undocumented functionality "zfs diff" not in the manual page for zfs, for example if I run "zfs ...
2012-03-12
-
- changeset: 13636:776c3d9cae69
tag: tip
user: Garrett D'Amore <garrett@damore.org>
date: Su... -
-
- The pcs driver is a completely empty shell. It has only a do-nothing attach/detach, and no additional functionality ...
-
01:20 PM Bug #1844: xen: "disk_link: invalid disk device number"
- Fixing patch information:
Parse Xen device number accordingly to Xen's commit
commit e951dcf324f7ce83d... -
- Illumos should support RFC 5006/6106 to be able to operate in IPv6-only environment with DHCPv6 only.
2012-03-11
-
- Resolved in r13634 commit:44267816c977
-
- Resolved in r13635 commit:022f1af9e21c
-
- Updated bug title.
-
- Correct. That package comes from the g11n consolidation. In fact, I'd love to see this code reimplemented properly ...
-
- Implemented this requires updating /usr/lib/iconv/alias which comes from pkg:/system/library/iconv/utf-8 but that doe...
-
- Forget it. My mistake. nl_langinfo behaves exactly as it should.
-
- nl_langinfo(CODESET) should try to obtain encoding info from LC_ALL, LC_CTYPE and LANG in that order. Not just LC_CT...
-
- I have a fix. Not sure if it's completely sane but it seems to work. Will send for review later today after more te...
-
- It seems our version of iconv does not support "char" either.
-
2012-03-10
-
- I think I see where the problem is.
2012-03-09
-
-
- Currently V4L2 layer in Illumos is just to support few webcameras using USB video. It would be nice to have complete ...
-
- Some basic work done my Mikore Li, published here
http://hub.opensolaris.org/bin/view/Project+bluetooth/ -
- Add bluetooth support, at least for mouse
-
- Changing dump_bzip2_level has no effect because we don't use bzip2 in serial mode at all, because it's so expensive.
2012-03-08
-
- The appropriate idiom is, iirc:...
-
- If we're doing this, we may want in.mpathd to do the right thing
-
- This limits the number of ipmp interfaces that can be created. ifconfig fails with:
ifconfig: cannot ping in.mpath... -
- As Garrett suggested, it is time to remove this service and just deliver editable minimal device.tab and dgroup.tab
... -
- /usr/gnu/bin/printf doesn't have this problem, so I guess illumos printf shoud be fixed somehow
-
- Actually, standard does not say explicitly about *input*. In any case this utility (printf) is crazy since it already...
-
- Igor Pashev wrote:
> Andrew Stormont wrote:
> > That's not right. LC_NUMERIC should effect the output of printf: h... -
- At least on SPARC, md(7D) still uses _depends_on, and DEBUG kernels complain.
It should stop.
2012-03-07
-
- Andrew Stormont wrote:
> That's not right. LC_NUMERIC should effect the output of printf: http://pubs.opengroup.org... -
- That's not right. LC_NUMERIC should effect the output of printf: http://pubs.opengroup.org/onlinepubs/007904975/util...
-
- /bin/printf fails with floating point argument in non-C locale:
~$ locale
LANG=C
LC_CTYPE=C
LC_NUMERIC=C
LC_TIME... -
- Most programmes ship message catalogs in directories named after language code, without country code or encoding (e. ...
-
-
- As a GCC extension, GCC used to define 'extern inline' to mean that a function was only ever emitted inline and no vi...
2012-03-06
-
- I have discovered two issues with the libbrand dtd's:
1: forcedlogin_cmd is optional but the dtd marks it as requi... -
- The patch makes libbrand print out xml parsing errors.
-
- open_brand fails with a rather generic looking error message if the config.xml file for your brand does not pass vali...
-
- Possibly related to the fix for #2083?
-
- I should mention I'm running a custom-built kernel. Here's the tree I built it from:
$ hg id -i
ab0e9bcbdf3e
It's... -
- My system at home panicked overnight:...
-
- Resolved in r13633 commit:7d8ad953b304
2012-03-05
-
- Yet more on the general theme: "GCC is fussy about pointer casting". I know you're excited!
There's a handful of... -
- When wsdiff is used, there is a set of spurious changes which one tends to have to memorize to know are not actual di...
-
- Reopening this as it was backed out due to 2221.
-
- I've done some deep analysis here, and the problem is that mpt is using the devinfo structure internals directly, mos...
-
- Seeing this on my ThinkPad T410 now with the wireless card removed.
2012-03-04
-
10:41 PM Feature #2239 (Resolved): Integrate the virtio net driver
- It's about time.
-
- upd:
console output:
panic[cpu3]/thread=ffffff04e66df440: BAD TRAP: type=e (#pf Page fault) rp=ffffff001f45... -
- During normal work, OI-148 server rebooted. The boot enviroment never booted again, freezes on startup.
Managed to b... -
- I don't believe so.
Just tried rolling back as before.
mailstats will work with : mailstats -f /etc/mail/stati... -
- I assume that /etc/mail/statistics is actually being updated.
The problem is that `mailstats' doesn't know about it.... -
- workaround is to add...
-
- Thought while trying to get sendmail firmed up that statistics would be nice.
From the docs, notably /etc/mail/cf/... -
- Resolved in r13632 commit:ab0e9bcbdf3e
2012-03-03
-
08:37 PM Feature #252: Update Fuse to 2.8.4
- Hi Gordon,
I certainly would like this, thanks.
Sheldon
Gordon Ross wrote:
> BTW, if someone _does_ want a F... -
- #998 was backed out which should clear this up.
2012-03-02
-
- The closed binaries contain CTF information which was made unique against a genunix from long ago. This has always b...
-
- Rich Lowe wrote:
> This is a list of things I would try, without any real science to them at present:
>
> - Boo... -
- From an email I sent earlier...
-
- This is a list of things I would try, without any real science to them at present:
- Boot with the IOMMU disable... -
- System panics with putback 13630 (#998 obsolete DMA driver interfaces should be removed) and closed mpt driver.
I ...
2012-03-01
-
Resolved in:
changeset: 13630:76b4d0b12a17
tag: tip
user: Garrett D'Amore <garrett@damore.org...-
10:20 PM Feature #2216 (Closed): add tool to check for/download updated CA certs
- CA certs are stored in PEM format under /etc/certs. We need a script to check for and download updates from a source ...
-
- Resolved in r13629 commit:741592a55c4c
-
- I've been hacking a little on this. Checking in with developer@ on porting strategy. Will carry it forward unless som...
-
01:26 PM Bug #2215 (New): all shell scripts should be checked if they don't inherit bad environment variables
- In this special case in some other OS using the same shell script I did run across the problem hat the script
illum...
2012-02-29
-
- Resolved in r13627 commit:81f67a2ecd98
-
- Resolved in r13628 commit:03b37ed99c1b
-
- Presuming we had to fix this, Alasdair asked "What about the closed drivers?".
Presuming that we made ddi_quiesce_... -
- A driver can prevent fastreboot in exactly one way: It "needs" to quiesce, yet fails to.
The specifics of "fails ... -
- Though the nature of should_implement_quiesce also suggests that this won't, even then, actually make us not fast reb...
-
- KVM is a driver which should_implement_quiesce() thinks would be safe regardless, so that explains that.
-
- https://github.com/richlowe/illumos-kvm/commit/e7e6d43e47451dff7ce60587eead8423df038df7
Is the cowards way out, ca... -
- Resolved in r13626 commit:f228617ce378
-
- Resolved in r13625 commit:2933dbe8e1a8
2012-02-28
-
- the kadmin GUI does this:...
-
- What do you mean by odd behavior? As for the primary itself, it's pretty easy to implement.
-
05:03 PM Bug #2208 (Resolved): Seems our find doesn't support the '-links' option, (as mandated by POSIX)
- Seems our find doesn't support the '-links' option, which is mandated by
POSIX: <http://pubs.opengroup.org/onlinepub... -
- Resolved in r13624 commit:7a2d5713360e
-
- I'm expecting that this was thus #1333
-
04:06 PM Bug #2129 (Feedback): interrupt increase on some Sandy Bridge CPUs
- I am happy to report that the issue does not occur with release 151a2. Please feel free to mark it closed.
-
04:12 AM Bug #2207 (New): iwp_load_run_firmware(): timeout waiting for run uCode load.
- I recently put a 6300 Intel wifi card into my T520. (It works fine on Windows 7.) I haven't been able to associate ...
2012-02-27
-
- garrett@velocity{19}> hg head
changeset: 13623:cd878e98cd0d
tag: tip
user: Dan McDonald <danmcd@n... -
- No-brainer of a change:...
-
- And I quote:...
-
-
- lm75(7D) allows setting of various temperatures, in a somewhat annoying format:
See http://www.datasheetcatalog.org/... -
- usr/src/lib/fm/topo/modules/common/ses/ses.c
on few places the code calls functions with return values but neither... -
- Few notes:
1) Solaris LDAP SDK (libldap.so.5) does not support StartTLS, but SSL is supported
2) but Apache apr... -
10:51 AM Feature #2190 (New): Apache (and PHP) compliation against OpenLDAP libraries
- Hi,
Current (Solaris) LDAP libraries are preventing use of LDAP functionality from both Apache (mod_ldap) and PHP ...
Also available in: Atom