Project

General

Profile

Activity

From 2013-01-09 to 2013-02-07

2013-02-07

10:06 PM Bug #3271 (Closed): ktkt_warn service is not renewing ticket
Not OI specific. Please submit under Illumos-specific bug reports. Ken Mays
09:52 PM Bug #3305 (Resolved): smbfs mount shows password
... Gordon Ross
09:45 PM Feature #3537 (Closed): want pool io kstats
From Matt Ahrens's bug report at Delphix:
We want a io kstat, which can be consumed by iostat(1m), which tells us ... Christopher Siden
06:36 PM Feature #3536 (New): mdb could use a larger input buffer
Several times when debugging the link-editor, I have been unable to :r it under mdb with the failing command line, be... Rich Lowe
05:06 PM Feature #3535 (Closed): mdb ::print should pretty-print ipaddr_t and friends
From Sebastien Roy's initial bug report at Delphix:... Christopher Siden
03:55 PM Bug #3534 (Resolved): Disable EEE support in igb for I350
The energy efficient Ethernet (EEE) support in Intel's I350 GigE NIC drops link on directly-attached link cases.
W... Dan McDonald

2013-02-06

11:43 PM Feature #3525: Persistent L2ARC
I've written a wiki post about the design at: http://wiki.illumos.org/display/illumos/Persistent+L2ARC Sašo Kiselkov
04:21 PM Feature #3525 (Feedback): Persistent L2ARC
Implementation is now feature-complete, new webrev at:
http://cr.illumos.org/~webrev/skiselkov/3525/ Sašo Kiselkov
11:42 PM Bug #3528: setting permission with chmod kills delete_child permission on owner acl
OK, I found the workspace where I left this fix parked.
The problem here, btw, is that Windows really expects the ... Gordon Ross
10:31 PM Bug #3528: setting permission with chmod kills delete_child permission on owner acl
This should be an illumos bug, btw. (I don't have the rights needed to fix that.)
We fixed this in NexentaStor, and... Gordon Ross
07:17 PM Bug #3533 (New): machrel.{intel,amd}.c aren't safe with strict alignment
In many places in machrel.intel.c and machrel.amd.c you see LINTED comments, this is to gag possible alignment proble... Rich Lowe
02:55 PM Bug #3532 (New): Aborted zfs recv causes NULL pointer kernel panic
Aborting a zfs recv operation is causing the system to panic. This was easily reproduced, 5 out of 5 times. Running i... Patrick Domack
02:25 PM Bug #3531: Race between log_sysevent_filename() and log_event_upcall() can cause panic
*Fix Testing:*
For the fix testing I used this event generator (source for the /var/tmp/sysevent/sysevent binary i... Marcel Telka
02:22 PM Bug #3531: Race between log_sysevent_filename() and log_event_upcall() can cause panic
*HOW TO REPRODUCE THE ISSUE*
*Prerequisites:*
To reproduce these issues (on an OpenIndiana box) you need to mod... Marcel Telka
02:20 PM Bug #3531: Race between log_sysevent_filename() and log_event_upcall() can cause panic
*Root Cause 3:*
Another race we need to consider during the bugfix:... Marcel Telka
02:19 PM Bug #3531: Race between log_sysevent_filename() and log_event_upcall() can cause panic
The code in log_sysevent_filename() and log_event_upcall() is fragile. The another panic is possible (I reproduced it... Marcel Telka
02:12 PM Bug #3531: Race between log_sysevent_filename() and log_event_upcall() can cause panic
*Root Cause 1:*
In case the log_event_deliver() is running and it is trying to deliver events to syseventd and in ... Marcel Telka
02:05 PM Bug #3531 (Resolved): Race between log_sysevent_filename() and log_event_upcall() can cause panic
If the log_sysevent_filename() and log_event_upcall() are running concurrently it can cause several issues:
- two th... Marcel Telka
09:43 AM Bug #3530 (Resolved): ::smbsess prints wrong IP addresses in verbose mode, has no idea about IPv6
IPv4 used:... Yuri Pankov
06:05 AM Feature #3529 (Closed): iostat should display time used by dtrace
From Matt Ahrens's feature request at Delphix:... Christopher Siden

2013-02-05

07:16 PM Feature #3507 (Closed): Tunable to allow block allocation even on degraded vdevs
... Christopher Siden
07:09 PM Bug #3519 (Resolved): DTrace fails to resolve const types from fbt
... Christopher Siden
07:08 PM Bug #3520 (Resolved): dtrace internal error -- token type 316 is not a valid D compilation token
... Christopher Siden
07:08 PM Bug #3521 (Closed): clean up dtrace unit tests
... Christopher Siden
06:40 PM Bug #3492 (Resolved): some e1000g devices don't support 15 unicast addresses
Resolved in:... Yuri Pankov
11:24 AM Bug #3517: importing pool with autoreplace=on and "hole" vdevs crashes syseventd
For the sake of completeness:... Yuri Pankov
10:57 AM Bug #3528 (Closed): setting permission with chmod kills delete_child permission on owner acl
Hi,
we do have the following problem:
when a directory has the delete_child permission, after changing mode bit... Klaus Steinberger
07:17 AM Bug #3435 (Closed): nfssrv causes bad mutex crash
Patrick Domack wrote:
> Ah, that is the case, I had used the posted patch, but the attached patch is different.
I... Marcel Telka
03:36 AM Bug #3435 (Feedback): nfssrv causes bad mutex crash
Ah, that is the case, I had used the posted patch, but the attached patch is different. Patrick Domack

2013-02-04

10:54 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
Fortunately and unfortunately no... I will post as soon as an issue happens.
Cheers,
Dave
 David Scharbach
05:28 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
David, have you seen this bug crop up again? Just checking in so it doesn't get forgotten. Sašo Kiselkov
08:22 PM Bug #3435 (Closed): nfssrv causes bad mutex crash
*Analysis of patrickdk.com/crash3.tar.gz*... Marcel Telka
12:37 PM Bug #3435: nfssrv causes bad mutex crash
patrickdk.com/unix.1.gz
Ok, here is a another crash.
patrickdk.com/crash3.tar.gz... Patrick Domack
10:50 AM Bug #3435: nfssrv causes bad mutex crash
The http://patrickdk.com/crash2.tar.gz contains these files:... Marcel Telka
03:49 PM Feature #3525 (Closed): Persistent L2ARC
This feature implements a light-weight persistent L2ARC metadata structure that allows L2ARC contents to be recovered... Sašo Kiselkov

2013-02-03

05:07 AM Bug #3523: SMB NT Notify returning too soon
The implementation of NT Notify uses flags in the node structure (node->flags,
and see NODE_FLAGS_NOTIFY_CHANGE) whic... Gordon Ross
03:51 AM Bug #3523 (Resolved): SMB NT Notify returning too soon
When many Windows clients have an "explorer" window open on some directory,
where they all have an SMB NT Notify ope... Gordon Ross
05:06 AM Bug #3223: SMB FindFirst/FindNext don't fill in Last Name Offset
oops, wrong issue. Gordon Ross

2013-02-02

11:17 PM Feature #3522 (Closed): zfs module should not allow uninitialized variables
The initial fix for '3236 zio nop-write' had a bug in it caused by uninitialized variables that could have been avoid... Christopher Siden
08:36 PM Bug #3521 (Closed): clean up dtrace unit tests
Adam Leventhal did some cleanup in the following dtrace unit tests:... Christopher Siden
04:47 AM Bug #3514 (Resolved): Spurious errors setting up new SMB client sessions
... Gordon Ross
12:41 AM Bug #3435: nfssrv causes bad mutex crash
TIME UUID SUNW-MSG-ID
Feb 01 2013 19:34:26.860243000 31798... Patrick Domack
12:30 AM Bug #3520: dtrace internal error -- token type 316 is not a valid D compilation token
The trace() and print() actions need to explicitly handle the case of having an aggregation for an argument. Adam Leventhal
12:29 AM Bug #3519: DTrace fails to resolve const types from fbt
Looking at the arc_read_nolock fuction, we can see that the types for which DTrace is struggling are the const ones:
... Adam Leventhal

2013-02-01

11:23 PM Bug #3520 (Pending RTI): dtrace internal error -- token type 316 is not a valid D compilation token
Christopher Siden
10:37 PM Bug #3520 (Resolved): dtrace internal error -- token type 316 is not a valid D compilation token
If you pass an aggregation to the trace() or print() actions dtrace fails due to an internal error. ... Christopher Siden
11:23 PM Bug #3519 (Pending RTI): DTrace fails to resolve const types from fbt
Yeah, I just sent a review to the list. Christopher Siden
11:22 PM Bug #3519: DTrace fails to resolve const types from fbt
Is this one of the issues you guys have a fix for already? Rich Lowe
10:36 PM Bug #3519 (Resolved): DTrace fails to resolve const types from fbt
Example:... Christopher Siden
10:19 PM Bug #3518: Incorrect hostid calculation
Here's the fix for my bug. Rather than switching to unsigned values throughout, I think masking off the sign bit is ... Garrett D'Amore
10:13 PM Bug #3518 (Resolved): Incorrect hostid calculation
During some additional testing on some other hardware, we found a sign extension bug that I introduced when I committ... Garrett D'Amore
10:12 PM Bug #3498 (Resolved): panic in arc_read(): !refcount_is_zero(&pbuf->b_hdr->b_refcnt)
... Christopher Siden
09:29 PM Bug #3435: nfssrv causes bad mutex crash
Please provide the crash dump file. Thanks. Marcel Telka
09:07 PM Bug #3435 (Feedback): nfssrv causes bad mutex crash
Reopened per submitter's request. Yuri Pankov
09:05 PM Bug #3435: nfssrv causes bad mutex crash
I've been running the patch above for alittle over a week, and it just crashed again, the exact same way.
TIME ... Patrick Domack
09:22 PM Bug #3333: USCSI ioctl: SCSI passthrough of ATA command: "0xe5: check power mode" wakes up sleeping drive
I've tested this under Openindiana and OmniOS. Thomas Geppert
09:01 PM Bug #3517 (Closed): importing pool with autoreplace=on and "hole" vdevs crashes syseventd
spa_check_removed() uses vdev_is_dead(), which does a bit of mis-service (for this case) by returning true for the "h... Yuri Pankov
08:28 PM Bug #3516 (Resolved): Leftover multidata transmit kernel statistics in ipv6
The ipv4 equivalents got removed in this commit:bd670b35a010421b6e1a5536c34453a827007c81
* ip6_frag_mdt_pkt_out
*... David Höppner
04:17 PM Bug #3515: sd gives RMW warnings for reads
webrev: http://cr.illumos.org/~webrev/hans/illumos-3515-webrev/
 Hans Rosenfeld
04:09 PM Bug #3515 (Resolved): sd gives RMW warnings for reads
The code to handle short or misaligned I/O in the sd driver does not distinguish between reads and writes when decidi... Hans Rosenfeld
04:40 AM Bug #3514 (Pending RTI): Spurious errors setting up new SMB client sessions
Gordon Ross
12:28 AM Bug #3514: Spurious errors setting up new SMB client sessions
The way connection setup works in our SMB client is best explained by looking at:
http://src.illumos.org/source/xref/... Gordon Ross
12:24 AM Bug #3514 (Resolved): Spurious errors setting up new SMB client sessions
When the smb server has multiple threads asking for a session to the AD server,
those threads will often ask at near... Gordon Ross
02:36 AM Bug #3508 (Resolved): NFS4 data corruption (client side)
... Gordon Ross
12:06 AM Bug #3511 (Resolved): dtrace.c erroneously checks for memory alignment on amd64
... Christopher Siden

2013-01-31

10:10 PM Feature #3500: Support LSI SAS2008 (Falcon) Skinny FW for mr_sas(7D)
You should add the IBM ServeRAID M1115 too, it's an almost identical successor of the M1015.
The sole difference betw... Mathieu Simon
07:39 PM Feature #3500: Support LSI SAS2008 (Falcon) Skinny FW for mr_sas(7D)
Changing the bug summary to be more accurate. I have made progress here too, BTW. Dan McDonald
07:45 PM Bug #3325 (Resolved): nsmb_close locking and teardown deadlock
... Gordon Ross
06:11 PM Bug #3325: nsmb_close locking and teardown deadlock
The smbiod is in the middle of shutting down it's connection to the AD server, typically due to it being idle. Durin... Gordon Ross
07:35 PM Bug #3513 (Closed): mismatch between SA header size and layout
This is a port of a ZoL fix. From their description:... Christopher Siden
07:28 PM Bug #3512 (Closed): rounding discrepancy in sa_find_sizes()
This is a port of a ZoL fix. From their description:... Christopher Siden
05:44 PM Feature #2832: libsocket should go the way of libpthread
Keith Wesolowski wrote:
> Apparently libsocket is already there, so this applies to libnsl only.
Sadly, it doesn'... Yuri Pankov
05:02 PM Bug #3510 (Feedback): libwrap 64 bit build failed
Right, if this is an RFE for a 64bit libwrap, you should update it so it makes that clear.
The fact you've tried... Rich Lowe
04:37 AM Bug #3510: libwrap 64 bit build failed
There doesn't seem to be an amd64 target for libwrap in illumos-gate. Yuri Pankov
04:30 AM Bug #3510 (Closed): libwrap 64 bit build failed
... Igor Kozhukhov
04:12 PM Bug #3468 (Resolved): mdb enhancements for zfs development
... Christopher Siden
04:12 PM Bug #3473 (Closed): mdb_get_module() returns wrong module
... Christopher Siden
04:11 PM Bug #3467 (Resolved): mdb_ctf_vread() could be more useful
... Christopher Siden
04:11 PM Bug #3470 (Resolved): ::whatis does not print callers from KMF_LITE
... Christopher Siden
04:10 PM Bug #3466 (Resolved): ::tsd should handle missing/NULL values better
... Christopher Siden
04:09 PM Bug #3465 (Resolved): ::walk ... | ::<dcmd> misinterprets input as symbol names
... Christopher Siden
04:08 PM Bug #3425 (Closed): first write to a new zvol can fail with EFBIG
... Christopher Siden
04:08 PM Bug #3422 (Closed): zpool create/syseventd race yield non-importable pool
... Christopher Siden
03:59 PM Bug #3511 (Resolved): dtrace.c erroneously checks for memory alignment on amd64
From Sebastien Roy's bug report at Delphix:... Christopher Siden
08:48 AM Bug #3496 (In Progress): sysevent runtime files are under /var/run
Marcel Telka
07:04 AM Bug #3431 (In Progress): sharectl prints garbage when it failed
Marcel Telka
02:27 AM Bug #3380 (Resolved): zfs man page: documentation for zfs allow is confusing
... Gordon Ross
02:26 AM Bug #3371 (Resolved): dumpadm -z is not documented in manpage
... Gordon Ross
01:24 AM Bug #3509 (Resolved): igb should not set alternate mac address on newer models
Robert Mustacchi
01:24 AM Bug #3509: igb should not set alternate mac address on newer models
Resolved in 7575615f3f759f94e5ff0b6ee3a4940dc9968c47. Robert Mustacchi

2013-01-30

07:02 PM Bug #3237 (Resolved): r10 is added 2 times to panic info
Garrett D'Amore
07:01 PM Bug #3237: r10 is added 2 times to panic info
commit ff512403afa294019aee0a9c4e4f778f1d7ed207
Author: Garrett D'Amore <garrett@dey-sys.com>
Date: Wed Jan 30 10... Garrett D'Amore
12:53 PM Bug #3237 (Pending RTI): r10 is added 2 times to panic info
Marcel Telka
07:00 PM Bug #3196 (Resolved): ao_gendisp use by make is racey
commit a0900badd8eb6bcca99411c8906b483e84f2da91
Author: Garrett D'Amore <garrett@dey-sys.com>
Date: Wed Jan 30 10... Garrett D'Amore
12:53 PM Bug #3196 (Pending RTI): ao_gendisp use by make is racey
Marcel Telka
06:40 PM Bug #3509 (Resolved): igb should not set alternate mac address on newer models
Currently igb always tries and sets the alternate mac address. However, since the 82580 it should not do this and in... Robert Mustacchi
05:17 PM Feature #3501 (In Progress): mdb: ::thread -d needs the SWITCH column
Marcel Telka
03:18 PM Bug #3503 (In Progress): usba doesn't initialize cc_no_resources, causes gibberish output in kstat:
David Höppner
02:02 PM Bug #3508 (Resolved): NFS4 data corruption (client side)
When writing to a file, from time to time some data don't make it to the server. Instead, a hole is left in the file.... Arne Jansen
09:00 AM Bug #1927 (In Progress): nfs4 stale clientid loop (OI server, Linux client)
The #1927 is not the same issue as #417. It is a combination of these two bugs:
* first is #417 (already fixed in ... Marcel Telka
03:42 AM Bug #1097: glob(3c) needs to support non-POSIX options
One more change, this one to supply the remaining buffer length to mbtowc() instead of a fixed value. See the file `... Gary Mills

2013-01-29

07:14 PM Feature #3507: Tunable to allow block allocation even on degraded vdevs
Webrev available at: http://cr.illumos.org/~webrev/skiselkov/3507/ Sašo Kiselkov
07:12 PM Feature #3507 (Closed): Tunable to allow block allocation even on degraded vdevs
This patch introduces a new boolean tunable named "zfs_write_to_degraded" which allows administrators to override the... Sašo Kiselkov
08:54 AM Bug #3196: ao_gendisp use by make is racey
*Root Cause:*
The root cause seems to be that the usr/src/uts/Makefile runs all intel related architectures in par... Marcel Telka
08:39 AM Bug #3196: ao_gendisp use by make is racey
I saw this in my build log:... Marcel Telka
08:54 AM Feature #3506 (Resolved): Use "hypervisor" CPUID bit to detect hypervisor environment
Almost all modern hypervisors set "hypervisor" CPUID bit, use it to more reliably detect the hypervisor environment, ... Yuri Pankov
05:01 AM Bug #3505 (Resolved): Creating LU unconditionally enables write cache on backing store device
Using `stmfadm create-lu` unconditionally enables write cache on backing store device - the code in usr/src/uts/commo... Yuri Pankov

2013-01-28

08:42 PM Bug #3502 (Resolved): panic in smb_server_shutdown (sl_count assertion)
... Gordon Ross
08:00 PM Bug #1097: glob(3c) needs to support non-POSIX options
Here are some more changes to glob.c, in response to reviewer's comments. The file `glob.c-rev.diff' contains these ... Gary Mills
07:02 PM Feature #604 (Resolved): CIFS server should traverse child mount points
... Gordon Ross
01:43 PM Bug #3449: System panics after zfs rollback
My system seems to have stablilized also after adding this patch.
Well it stabilized for a little over a week, and m... Patrick Domack

2013-01-27

06:16 PM Bug #3503 (Resolved): usba doesn't initialize cc_no_resources, causes gibberish output in kstat:
Rich:... Rich Lowe
12:39 AM Bug #3502: panic in smb_server_shutdown (sl_count assertion)
The panic is at smb_session.c:727... Gordon Ross
12:36 AM Bug #3502 (Resolved): panic in smb_server_shutdown (sl_count assertion)
This can happen if you disable and then quickly re-enable the smb service.
Here's what the stack looks like:
<pre... Gordon Ross

2013-01-26

07:18 PM Bug #749 (Resolved): "/usr/bin/kstat" should be rewritten in C
commit 66448911ac89c254e89afc52e114402115a497e3
Author: David Höppner <0xffea@gmail.com>
Date: Wed Jan 9 12:18:53... Garrett D'Amore
01:05 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
I am sorry I am such a n00b but I cannot answer if I do have the fixes from #1761. I have installed all of the updat... David Scharbach

2013-01-25

08:44 PM Bug #3422: zpool create/syseventd race yield non-importable pool
This is a regression caused by bug #3102. George Wilson's analysis:... Christopher Siden
08:40 PM Bug #3116 (In Progress): zpool reguid may log negative guids to internal SPA history
George Wilson
08:27 PM Bug #3425 (In Progress): first write to a new zvol can fail with EFBIG
We are over estimating the usage because we're using the wrong indirect block size. I have a fix that will be sent ou... George Wilson
08:06 PM Bug #3485 (Resolved): statd is not handling incoming SM_NOTIFY properly when the hostname is not resolvable
Christopher Siden
08:01 PM Bug #3485: statd is not handling incoming SM_NOTIFY properly when the hostname is not resolvable
... Christopher Siden
09:18 AM Bug #3485 (Pending RTI): statd is not handling incoming SM_NOTIFY properly when the hostname is not resolvable
Marcel Telka
03:13 PM Feature #3500: Support LSI SAS2008 (Falcon) Skinny FW for mr_sas(7D)
Actually, Ken, for Skinny, it's going to be mr_sas, not mpt_sas. The Skinny Firmware is closer to 2108/2208 firmware... Dan McDonald
01:41 PM Feature #3500: Support LSI SAS2008 (Falcon) Skinny FW for mr_sas(7D)
Thanks, Dan. Since imr_sas is currently unsupported and unmaintainable from a FOSS project perspective, it makes sens... Ken Mays
02:10 PM Bug #3036 (Feedback): Update hwdata to current upstream data
Ken Mays
11:09 AM Feature #3501 (Resolved): mdb: ::thread -d needs the SWITCH column
It would be really helpful to have the SWITCH column (see ::cpuinfo) in the ::thread -d output. Now, when you want to... Marcel Telka
09:18 AM Bug #3371 (Pending RTI): dumpadm -z is not documented in manpage
Marcel Telka
09:18 AM Bug #3380 (Pending RTI): zfs man page: documentation for zfs allow is confusing
Marcel Telka
07:55 AM Bug #1921: pstop fails if process is being truss-ed
PGRAB_FORCE allows to pass EBUSY for /proc/xxx/as - shouldn't we remove this flag here? Alexander Eremin

2013-01-24

08:39 PM Feature #2034 (Closed): Rewrite perl kstat in c
Closing as dup of 749. Garrett D'Amore
08:16 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
I'm uploading a copy of the dump to beast.ma. It'll be ~trisk/vmdump.3489 on beast.
Alternatively, http://trisk.dead... Albert Lee
07:27 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
There was a somewhat similar panic in #1761 - does your system have that fix?
Or this could be some other case of a ... Gordon Ross
07:30 PM Bug #3449: System panics after zfs rollback
Common approach to apply - "patch -p1 {PATCH}"
Building is described at "HOW_TO_BUILD illumos"
 Vitaliy Gusev
06:31 PM Bug #3449: System panics after zfs rollback
Is there any documentation on how to apply and build this patch? Andrew Rokitka
02:46 PM Feature #3500: Support LSI SAS2008 (Falcon) Skinny FW for mr_sas(7D)
This is also known as "Skinny MegaRAID". Earlier this week someone from the community granted me access to some skin... Dan McDonald
02:44 PM Feature #3500 (Resolved): Support LSI SAS2008 (Falcon) Skinny FW for mr_sas(7D)
Add support for LSI MegaRAID SAS 2008 [Falcon] hardware to the mpt_sas driver:
Products: Dell PERC H310, LSI 9220-... Ken Mays

2013-01-23

11:01 PM Bug #3498 (Resolved): panic in arc_read(): !refcount_is_zero(&pbuf->b_hdr->b_refcnt)
This is a panic that only occurs on debug builds, although on non-debug builds this might cause a deadlock. From Geor... Christopher Siden
10:05 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
Added gwr as a watcher, as I suspect he might be interested. Albert Lee
12:26 PM Bug #3493: Extra registers in SPARCv9 sendsig never used
Rich Lowe wrote:
> Out of interest: What's wrong with our AVX support?
Just incomplete (procfs, mdb) PSARC/2010/3... David Höppner
08:22 AM Bug #1588 (Resolved): nfs4 mirror mount hang
Marcel Telka

2013-01-22

04:43 PM Bug #1983: sshd problems
We had this problem when setting primarycache and secondarycache to
'metadata' for the zfs root file system on small... Udo Grabowski
03:18 AM Bug #1983: sshd problems

i see this problem within zones but not on the hosts themselves. i can make access available to these zones if a re... Jason Matthews
12:38 PM Bug #3496 (Resolved): sysevent runtime files are under /var/run
In 2008 this was implemented:
6711375 sysevent runtime files should be under /var/run
This changed several syse... Marcel Telka
11:16 AM Feature #3495 (New): Ralink 539x Wireless Devices Support in ral
Missing support for Ralink 539x devices (eg. 539b/5931) in ral(7D).
On Linux, that support is came within 3.7 rele... Luca De Pandis

2013-01-21

11:52 PM Bug #3494 (New): cross linkers probably report bad error messages
doreloc.c includes machdep.h, which includes the right machdep.h based on the platform preprocessor symbol (__i386, s... Rich Lowe
10:35 PM Feature #3419 (Resolved): usbftdi needs to support the BeagleBone
Resolved in 452fab4 Rich Lowe
06:44 PM Bug #3493: Extra registers in SPARCv9 sendsig never used
Out of interest: What's wrong with our AVX support? Rich Lowe
11:12 AM Bug #3493 (New): Extra registers in SPARCv9 sendsig never used
While working on AVX userland support i noticed that
sendsig for SPARCv9 never sets xregs_size. Because
xregs_size ... David Höppner
06:21 PM Bug #3462 (Closed): Lenovo T520 is hanging on power cord connect/disconnect
Still not reproducible. Closing. Marcel Telka
04:29 PM Bug #3476: Contribute Nexenta mpt_sas(7D) mods back to Illumos
Don't worry about reproducing 2430 for now. If I need help later, however, I'll let you know. Thanks! Dan McDonald
04:16 PM Bug #3476: Contribute Nexenta mpt_sas(7D) mods back to Illumos
I haven't actually had 2430 recur at all in some time, so I'm not sure I can test whether this fixes it or not.
I ... Rich Ercolani
03:47 PM Bug #3492: some e1000g devices don't support 15 unicast addresses
webrev: http://cr.illumos.org/~webrev/hans/illumos-3492-webrev/ Hans Rosenfeld
04:34 AM Bug #3492 (Resolved): some e1000g devices don't support 15 unicast addresses
The default maximum number of unicast addresses is 15 for e1000g. The driver allows you to set as many unicast addres... Hans Rosenfeld
06:24 AM Bug #3368 (Closed): NFSv4 and slow OS X Mountain Lion (10.8.2) Client
Okay. I'm closing this. Once you have some info showing what can we do from the OI/illumos side then please reopen th... Marcel Telka
01:33 AM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
I do not use ECC. I ran 3 passes on memtest86 and everything passed with flying colours.
I will capture the next ... David Scharbach
12:07 AM Bug #3489 (Feedback): panic in smbsrv:smb_mbc_vdecodef
Setting to feedback, this needs more info to correctly assess what's going on. So far it doesn't make much sense to me. Sašo Kiselkov

2013-01-20

11:51 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
After a day of looking at the issue, I feel I have absolutely no idea what's going on here. From the looks of it, %r1... Sašo Kiselkov
09:21 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
Scrap my previous tangent on the "leal" causing a memory fault, it just does arithmetic. Since %r12 = 0x77, after the... Sašo Kiselkov
05:06 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
Just a quick follow-up, the exact ::stack (with line numbers where I know them) to this is:
smb_mbc_vdecodef+0xb3(ff... Sašo Kiselkov
05:01 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
The bug occured at smb_mbc_vdecodef+0xb3, which, if I read the assembly correctly, corresponds to smb_mbuf_marshallin... Sašo Kiselkov
03:38 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
Here is some of the information on the mailing list discussion from my server.
$ fmdump -m
SUNW-MSG-ID: SUNOS-800... David Scharbach
03:32 PM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
File sent to Saso.
Cheers, David Scharbach
02:15 AM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
I can send a secure file link to those who need the dump file.
And could one of the admins change the name of this... David Scharbach
12:22 AM Bug #3489: panic in smbsrv:smb_mbc_vdecodef
This appears to be an internal problem in the smbsrv module (the in-kernel smb server). The failure is a bad trap (me... Sašo Kiselkov
01:36 PM Feature #644: Add Toshiba AL10/12/* devices to mpxio sym list
updated file Richard Elling
01:57 AM Bug #3368: NFSv4 and slow OS X Mountain Lion (10.8.2) Client
Marcel, I just wanted to say thanks for looking into this from the OI side.
We no longer have an active ticket wit... Jonathan Mullen
12:39 AM Bug #3368 (Feedback): NFSv4 and slow OS X Mountain Lion (10.8.2) Client
Looking into the 373004549_apple_support.tar.gz file...
Assuming the 129.15.78.217 is the MacOS NFS client and 129... Marcel Telka
01:31 AM Feature #3409: mdb: Open source implementation of nfs module
http://wiki.illumos.org/display/illumos/mdb%3A+Open+source+implementation+of+nfs+module Marcel Telka
01:30 AM Feature #3409 (In Progress): mdb: Open source implementation of nfs module
Marcel Telka
01:27 AM Bug #3472 (Closed): nfs causing kernel panic
This is a duplicate of bug #3449. Closing. Marcel Telka

2013-01-19

10:35 PM Bug #3489 (Closed): panic in smbsrv:smb_mbc_vdecodef
My OI system is hanging and requires a hard boot. This happens between 1 and 20 days. Since October, the system has... David Scharbach
09:11 PM Bug #2998 (Feedback): AHCI reports many "Illegal Request" with Intel C602 chipset
Bob Friesenhahn
09:09 PM Bug #2998 (In Progress): AHCI reports many "Illegal Request" with Intel C602 chipset
Bob Friesenhahn
09:02 PM Bug #2998: AHCI reports many "Illegal Request" with Intel C602 chipset
With OpenIndiana oi_151a7 I am only seeing a tiny number of illegal requests (<= 10) so I am closing this report. Bob Friesenhahn
05:03 AM Bug #3486 (New): SPARC ld_mach_eflags should initialize firstpass
The SPARC ld_mach_eflags uses 'firstpass' to control its behaviour, indicating whether this is the first object seen ... Rich Lowe
03:22 AM Bug #3485: statd is not handling incoming SM_NOTIFY properly when the hostname is not resolvable
*Root cause:*
In case the client's hostname is not resolvable at the NFS server _and_ the client's statd will send t... Marcel Telka
03:11 AM Bug #3485 (In Progress): statd is not handling incoming SM_NOTIFY properly when the hostname is not resolvable
Marcel Telka
03:10 AM Bug #3485 (Resolved): statd is not handling incoming SM_NOTIFY properly when the hostname is not resolvable
*Steps to reproduce:*
Prerequisites:
* illumos based NFS server named SERVER
* an NFS client with hostname CLI... Marcel Telka

2013-01-18

11:08 PM Feature #3484: enhance and document tail follow support
webrev: http://cr.illumos.org/~webrev/jclulow/3484/
(NB: the Author of these changes is _Bryan Cantrill <bryan@joyen... Joshua M. Clulow
09:34 PM Feature #3484 (New): enhance and document tail follow support
Currently, our *tail -f* does not pick up a truncation of the file. That is, if you're doing a *tail -f* of */tmp/fo... Joshua M. Clulow
06:11 PM Bug #3477 (Resolved): SunSSH config should accept TCPKeepAlive as synonym for KeepAlive
Resolved in 33c41f2b7f5e60517664bd7f392221865f833526. Robert Mustacchi
03:57 PM Bug #3449: System panics after zfs rollback
We had the system panic again before I could get the patch in place, but it was without an ESXi client in place. Ther... Greg Mason
03:02 PM Bug #3482 (Closed): tmpfs in a non-global zone does not abide by tmpfs_minfree
We have seen cases recently of gsort in non-global zones dying under somewhat odd conditions. In all cases, we died ... Bryan Cantrill
12:35 PM Bug #3481 (New): sbdadm RBAC missing privilege "file_dac_read"
'File System Management' RBAC profile is missing privileges for sbdadm command:... Piotr Jasiukajtis
12:44 AM Feature #452 (Resolved): getconf should support NPROCESSORS_*
Resolved in 3c1e517 Rich Lowe

2013-01-17

11:23 PM Bug #3477: SunSSH config should accept TCPKeepAlive as synonym for KeepAlive
webrev: http://cr.illumos.org/~webrev/jclulow/3477/ Joshua M. Clulow
11:10 PM Bug #3477 (Resolved): SunSSH config should accept TCPKeepAlive as synonym for KeepAlive
SunSSH borks if you have *TCPKeepAlive yes* in your *~/.ssh/config*:... Joshua M. Clulow
08:56 PM Bug #3474 (Resolved): tar should support -C on extract
Resolved in c92bf99ee60c0b1fe1289fef2b4034126c7867a5. Robert Mustacchi
11:26 AM Bug #3474: tar should support -C on extract
GNU tar allows multiple -C and uses the lastest one. Igor Pashev
01:02 AM Bug #3474: tar should support -C on extract
webrev: http://cr.illumos.org/~webrev/jclulow/3474/ Joshua M. Clulow
12:41 AM Bug #3474 (Resolved): tar should support -C on extract
Currently our tar supports a *-C* on creation and a few other operations, but does not support *-C* on extract like G... Joshua M. Clulow
07:22 PM Bug #3476 (Feedback): Contribute Nexenta mpt_sas(7D) mods back to Illumos
The Nexenta version of mpt_sas(7D) has several bugfixes and one misfeature backout that currently aren't in illumos-g... Dan McDonald
07:14 PM Bug #1773 (Resolved): LSI 9200 SAS MPT driver forces synchronous writes to device
Closing this as "resolved" since the filer found a solution to the issue. Dan McDonald
02:35 PM Bug #3475 (New): tar seems broke with E function modifier for long filenames with diacriticals
seems that there are increasing problems identified with solaris tar, in this case with the 'E' function modifier all... Richard PALO
05:41 AM Bug #3368: NFSv4 and slow OS X Mountain Lion (10.8.2) Client
UPDATE (not a good one):
I have been getting more emails from other users (read LARGE federal and research institu... Jonathan Mullen
12:25 AM Feature #3137: L2ARC compression
I've discovered a pretty huge bug in the way l2arc reads back zero-buffers (buffers squashed by compression to zero l... Sašo Kiselkov

2013-01-16

11:15 PM Feature #3035 (Closed): LZ4 compression support in ZFS and GRUB
... Christopher Siden
05:54 PM Feature #3137: L2ARC compression
sorry, i could have sworn I was seeing [L2]ARC...
The physical memory part is naturally targeting the ARC environme... Richard PALO
08:21 AM Feature #3137: L2ARC compression
Richard PALO wrote:
> Sašo Kiselkov wrote:
> > Reworked webrev at http://cr.illumos.org/~webrev/skiselkov/3137/ to ... Sašo Kiselkov
06:14 AM Feature #3137: L2ARC compression
Sašo Kiselkov wrote:
> Reworked webrev at http://cr.illumos.org/~webrev/skiselkov/3137/ to be based on the latest co... Richard PALO
12:12 AM Feature #3137: L2ARC compression
Reworked webrev at http://cr.illumos.org/~webrev/skiselkov/3137/ to be based on the latest committed LZ4 patch and re... Sašo Kiselkov
03:01 PM Bug #1756: iwp missing some Centrino 6205 IDs
Ryo Murakawa wrote:
> Albert Lee wrote:
> > It's possible this change is not sufficient for all 62xx devices. They se... Ryo Murakawa

2013-01-15

12:14 AM Bug #3473 (Closed): mdb_get_module() returns wrong module
From Matt Ahrens's bug report at Delphix:... Christopher Siden

2013-01-14

10:11 PM Bug #3445 (Resolved): ZFS tests for restricted aclmode
Resolved in 7b1753e Rich Lowe
10:11 PM Bug #3444 (Resolved): README for usr/src/test
Resolved in 7b1753e Rich Lowe
06:23 PM Bug #3472 (Closed): nfs causing kernel panic
Kernel panic caused when automounting a stale nfs share from a remote machine. The share existed, but the directory ... Andrew Rokitka
05:31 AM Bug #3471 (New): cscope-fast writes constant strings when erroring
When erroring (most commonly) cscope-fast will write into constant strings, and crash instead.
the cscope putmsg()... Rich Lowe

2013-01-12

08:20 PM Bug #1097: glob(3c) needs to support non-POSIX options
One more small change, this one to replace wctomb() by iswascii() . See the attached file glob.c-asc.diff for detail... Gary Mills
06:20 PM Bug #3463 (Resolved): agent lwp clobbers amd64 abi stack redzone
Resolved in f6dcd367bb2ab63785ea4839111eb4846066404d. Robert Mustacchi
12:13 AM Bug #3469: dbuf_read_impl shows too much enthusiasm
So, the only time we've ever seen this was in a pool that was already essentially permanently damaged due to running ... Joshua M. Clulow

2013-01-11

09:23 PM Bug #3470 (Resolved): ::whatis does not print callers from KMF_LITE
From Matt Ahrens's original bug report at Delphix:... Christopher Siden
05:53 PM Bug #3442 (Resolved): sgs checks should be run in normal builds
Resolved in 9c5a3148 Rich Lowe
05:52 PM Bug #3451 (Resolved): archive libraries with no symbols shouldn't require a string table
Resolved in 87c7234 Rich Lowe
05:52 PM Bug #3450 (Resolved): elfdump crashes on non-Solaris/Linux objects
Resolved in 8a8ef76 Rich Lowe
05:52 PM Bug #3436 (Resolved): relocatable objects also need sloppy relocation
Resolved in 396f90f Rich Lowe
05:52 PM Bug #3439 (Resolved): discarded sections shouldn't end up on output lists
Resolved in c866e23 Rich Lowe
05:52 PM Bug #3453 (Resolved): GNU comdat redirection does exactly the wrong thing
Resolved in 84f7afe Rich Lowe
05:52 PM Bug #3441 (Resolved): -zassert-deflib messages are in the wrong place
Resolved in 059c3de Rich Lowe
05:52 PM Bug #3440 (Resolved): librtld_db demos need warnings gagged
Resolved in f8a567b Rich Lowe
03:23 PM Bug #3469: dbuf_read_impl shows too much enthusiasm
Dumb question perhaps, but will other parts of ZFS catch this corruption with your attached fixed? If not, should so... Jason King
07:21 AM Bug #3469: dbuf_read_impl shows too much enthusiasm
webrev: http://cr.illumos.org/~webrev/jclulow/3469/ Joshua M. Clulow
07:15 AM Bug #3469: dbuf_read_impl shows too much enthusiasm
After the heap corruption, *::kmem_verify* reported several pages of memory were destroyed:... Joshua M. Clulow
07:06 AM Bug #3469 (New): dbuf_read_impl shows too much enthusiasm
In the post-mortem for a zpool containing byzantine corruption (induced previously by #2638) it was discovered that i... Joshua M. Clulow
03:53 AM Bug #3468 (Resolved): mdb enhancements for zfs development
From Matt Ahrens's original bug report at Delphix:... Christopher Siden
03:52 AM Bug #3467 (Resolved): mdb_ctf_vread() could be more useful
From Matt Ahrens's original bug report at Delphix:... Christopher Siden
03:49 AM Bug #3466 (Resolved): ::tsd should handle missing/NULL values better
From Matt Ahrens's original review at Delphix:... Christopher Siden
03:25 AM Bug #3465 (Resolved): ::walk ... | ::<dcmd> misinterprets input as symbol names
From Matt Ahrens's original bug report at Delphix:... Christopher Siden
12:35 AM Bug #3464 (Closed): zfs synctask code needs restructuring
The locking code around dsl_datasets and dsl_dir's is responsible for several deadlocks and race conditions, most pro... Matthew Ahrens
12:22 AM Bug #3447 (Closed): improve the comment in txg.c
... Christopher Siden

2013-01-10

10:49 PM Bug #3462: Lenovo T520 is hanging on power cord connect/disconnect
I just found that after the BIOS update to version 1.41 I am no longer able to reproduce the issue. I'll test the mac... Marcel Telka
06:35 PM Bug #3462 (Closed): Lenovo T520 is hanging on power cord connect/disconnect
Lenovo ThinkPad T520 is hanging almost always on power cord connect/disconnect. Tested with oi_151a7.
*Workaround:... Marcel Telka
10:32 PM Bug #3463: agent lwp clobbers amd64 abi stack redzone
Webrev for fix is here: http://cr.illumos.org/~webrev/jclulow/3463/
Replacing */usr/lib/amd64/libproc.so.1* (via _... Joshua M. Clulow
10:30 PM Bug #3463 (Resolved): agent lwp clobbers amd64 abi stack redzone
_NB: this bug mirrors Joyent's OS-1804._
h2. Overview
Using *libproc*, a process may impose upon another proces... Joshua M. Clulow
02:23 PM Bug #3435: nfssrv causes bad mutex crash
*Root cause:*
In case the rfs3_lookup() is called with exi != NULL and the exi is not EX_PUBLIC and the nfs3_fhtov... Marcel Telka
02:16 PM Bug #3435 (Closed): nfssrv causes bad mutex crash
Marcel Telka
02:15 PM Bug #3435: nfssrv causes bad mutex crash
Even the #3449 is different (a bit) than this one, there are some common aspects: both issues were introduced by #298... Marcel Telka
06:01 AM Bug #3435: nfssrv causes bad mutex crash
Rich, are you talking about #3449? If so, then I believe this issue is different. Marcel Telka
04:07 AM Bug #3435: nfssrv causes bad mutex crash
Yeah, Dan had an analysis for that too, but I guess it didn't make the bug?
If you look at where it's freed, there... Rich Lowe
03:10 AM Bug #3435: nfssrv causes bad mutex crash
The problem is that we do have the already freed exportinfo in the exptable:... Marcel Telka

2013-01-09

04:44 AM Bug #3435: nfssrv causes bad mutex crash
The panic stack is:... Marcel Telka
12:54 AM Bug #3454 (New): SMF should log actions performed
SMF needs commands executed and reasons for execution logged for historical debugging. For example, given this manif... Nate Fitch
12:41 AM Bug #3453 (Resolved): GNU comdat redirection does exactly the wrong thing
In gnu_comdat_sym we attempt to determine whether a section name is that of a GNU comdat section, which are named '<s... Rich Lowe
 

Also available in: Atom