Activity
From 2014-04-08 to 2014-05-07
2014-05-07
-
10:58 PM Bug #4768 (Pending RTI): Handling of rfs4_start_time is fragile
-
03:52 PM Bug #4853: illumos-gate is not lint-clean when built with openssl 1.0
- Webrev is here: http://cr.illumos.org/~webrev/alp/4853/
-
- illumos-gate can be built against openssl 1.0 and works properly, but lint produces some warnings:...
-
- In addition to cl, we might also leak nc in _clnt_tli_create_timed().
Similar cl leak after the strdup() failure i... -
05:21 AM Bug #4851 (Closed): IPsec should use NANOSEC & friends from time.h
- "git commit cf0ee00a494fba00232790e34be9db9a41a516fd":https://github.com/illumos/illumos-gate/commit/cf0ee00a494fba00...
2014-05-06
-
- "git commit 92db401c04def311ec2fbd6cab722a35cd63f511":https://github.com/illumos/illumos-gate/commit/92db401c04def311...
-
-
- The rpc_clnt_auth(3nsl) man page contains this:...
-
06:23 PM Bug #4851 (Closed): IPsec should use NANOSEC & friends from time.h
- Subject pretty much describes it.
Ideally the new code should compile to identical code as the pre-cleanup source. -
- The IPsec part of this bug (gathering SPD information via DNSSEC) might be possible, but the IKE part is not, due to ...
-
- A similar change will be needed in ipsecah as well.
-
- There are two possible file descriptor leaks in tlicall().
*Leak 1*
Here is the source code:... -
11:08 AM Bug #4846: HAL partition names don't match real partition names
- Updated webrev: http://cr.illumos.org/~webrev/andy_js/4846-3/
Example output of eject -l before the changes:... -
-
05:09 AM Bug #4840 (Resolved): ddi_dma_mem_alloc()/ddi_dma_mem_free() are O(n**2)
- commit a7175e2030b8105e954880bdc63bf0e3dba62d2c
Author: Garrett D'Amore <garrett@damore.org>
Date: Fri May 2 21:3... -
04:47 AM Feature #4843: add ld --as-needed and --no-as-needed
- Apparently Or*cle Solaris has added this _compatibility_ feature
http://docs.oracle.com/cd/E36784_01/html/E36870/l...
2014-05-05
-
06:26 PM Bug #4849 (Closed): strlcpy(9F) points to the wrong header
- The manual page suggests to look in ...
-
- "git commit f93d2c191d5ef071436181338612f79b8daa751c":https://github.com/illumos/illumos-gate/commit/f93d2c191d5ef071...
-
11:21 AM Bug #1784 (Pending RTI): zone configuration passes zonecfg but not zoneadm for limitpriv property
-
- There is following piece of code in nb_send():...
-
- "git commit 14b7b4606a1faf7fc4c0bbcc2946782d6538850e":https://github.com/illumos/illumos-gate/commit/14b7b4606a1faf7f...
-
02:51 PM Bug #4813: WARNING: kb8042: inconsistent getinfo(9E) implementation
- Since I don't have a debug build BE handy, here's info from a non-debug build (very similar version)...
This was f... -
- "git commit d05b42383f649c1384ae2af878515cc2c6674ec6":https://github.com/illumos/illumos-gate/commit/d05b42383f649c13...
-
- For OmniOS, I defined PCFLAGS (Perl CFLAGS) in Makefile.targ. This one is slightly different from OmniOS, because Om...
-
- This is easy to fix as well.
Apply this patch to $SRC/cmd/perl/contrib/Sun/Solaris/Utils/Makefile:...
2014-05-04
-
- Example errors now displayed after changset:
Already mounted:... -
- Webrev: http://cr.illumos.org/~webrev/andy_js/4845/
-
-
- Both commands also fail to catch cases where the inserted media contains no mountable volumes.
-
- The rmmount or rmumount commands print out an error message unless the device they're trying to mount/unmount is alre...
-
- Webrev: http://cr.illumos.org/~webrev/andy_js/4846/
-
- For some reason that I cannot grok HAL is rewriting partition names for devices. The name /dev/dsk/c5t0d0p1 becomes ...
-
- This isn't a bug in ctfconvert. ctfconvert is telling you the truth, you have no dwarf in the file you built. Specifi...
-
05:00 PM Bug #4847 (Rejected): ERROR: ctfconvert: file does not contain valid DWARF data: Dwarf_Error is NULL
- ctfconvert has problem with 64bit build of libsqlite...
-
05:10 PM Bug #4453: driver/storage/sdcard package missing driver for sdcard
- BUT,
sometime, back in 151a7 it was working. That was on same hardware as today (/hipster).
so, one more functi... -
- Garrett D'Amore wrote:
> There is indeed *no* sdcard driver.
>
> Some time ago (years!) I rearchitected the sdcar... -
-
- http://cr.illumos.org/~webrev/risto3/il_4843/
-
- Feature request to add gnu ld synonyms `--as-needed` and `--no-as-needed`, respectively `-z ignore` and `-z record` ...
-
- Simon Klinkert wrote:
>
> We’ve encountered a side effect when dropping the initial delay from 1s to 1ms. The dist... -
- The actual locking at the NFSv4 server is implemented in setlock() - see file nfs4_srv.c.
In a case the file is al...
2014-05-03
-
- According the RFC 7530, section _9.4. Blocking Locks_, the NFSv4 server should maintain an ordered list of pending bl...
-
09:01 PM Bug #4400: kernel panic
- Keyboard hotplug events are because I'm using a 4 port KVM switch to access the system. I've got the coredump if yo...
-
- Do you have any further details about what you were doing when this occurred? I see lots of what look like keyboard ...
-
08:19 PM Bug #4826: ILB NAT has some serious problems with vmem arena
- The real problem (which cause I misinterpreted in 4826) was in incorrect order of resource release on IP stack shutdo...
-
- Btw, please don't rip out the C99isms localized int loop control variables. We have C99 in our kernel, and localizin...
-
- Dan, lets just go ahead and get these integrated.
-
- The word "function" should be _italic_, not *bold*, here in the lockf(3c) man page (three occurences):...
-
- If we yank it out of here, we should probably stick it into its own repo. I might spend a bit of effort setting that...
-
- What would be cool is to add some property that prevented attempts to set the dump device to something "non-useful". ...
-
- Doing some more reading, I've verified that indeed the TEAC model you found is a USB device.
The problem you may h... -
- There is indeed *no* sdcard driver.
Some time ago (years!) I rearchitected the sdcard stack so that it uses a more... -
- lms is a daemon that interfaces to heci, which provides local access to the AMT services on the system. Its never be...
-
- I just tested. You almost *CERTAINLY* forgot to add -D_KERNEL.
All kernel APIs (DDI) require the -D_KERNEL define... -
- 1ms back off may be under the threshold for user land sleep times (clock is often 100Hz, i.e. 10msec). I'd recommend...
-
- Please include the contents of "prtconf -vp" and "grep kb8042 /etc/path_to_inst"
The kb8042 driver assumes there i... -
- At Pluribus, we found that plumbing additional vnics on ixgbe (with many rings) could take a very long time. Some ti...
2014-05-02
-
- As Josef <jeffpc> pointed it is unconfirmed, based on wrong assumption.
Need still some investigations...
-
- Webrev:
http://mtc.md/sscdvp/webrevs/issue_4826/
Regards,
Serghei Samsi
-
09:26 PM Bug #4839 (Closed): 3294 generated elfdump struct layouts by hand, shouldn't have
- In elfdump, we have a tool that generates the structure layout definitions for us. 3294 didn't use it when adding pr...
-
- The results of post-4596's ip_srcid_find_id() are TRUE when there's a v4mapped/v6only MATCH between srcid and the des...
-
- In a case the NFSv4 client cannot get a lock, the exponential backoff is implemented in nfs4_block_and_wait(). The up...
-
01:09 AM Feature #4836 (New): implement processor_affinity(2) à la Solaris 11.2
- Illumos and Solaris have had *processor_bind(2)* for quite some time, but this only allows the programmer to bind a p...
2014-05-01
-
- There are few minor formatting issues in the getexecattr(3secdb) man page.
Extra space after paren (twice):
<pr... -
- Simple. Move the /usr/has stuff to a separate package.
-
- You are correct. I incorrectly labeled it as GNU file.
I wonder if it would make sense to just drop file from illu... -
- You are correct. I incorrectly labeled it as GNU file.
I wonder if it would make sense to just drop file from illu... -
- It has been deprecated and at the end of the day it's just another dependency on HAL and DBus.
-
- Pre-3900, the ON perl modules (which are 32-bit) were built with -DPERL_EUPXS_ALWAYS_EXPORT -D_LARGEFILE_SOURCE -D_FI...
-
- As part of the 3900 cleanup, the Makefile for Utils.so needs to include -lgen and -lintl, as the pre-3900 code did.
... -
- Nexenta has done this already: https://github.com/Nexenta/illumos-nexenta/commit/246aaaaf77f6c13cc11ebf0a928045f19990...
-
10:23 AM Bug #4830 (New): 'zfs diff' doesn't print headers
- Not sure if it should, but as -H option description mentions it ("Give more parsable tab-separated output, without he...
-
-
- "git commit dff8cdb7c81be0916631bfe5b4384516b430227f":https://github.com/illumos/illumos-gate/commit/dff8cdb7c81be091...
-
- "git commit 0900f4f20e4cdeb8e5af4ba32aca32d28aeba2f9":https://github.com/illumos/illumos-gate/commit/0900f4f20e4cdeb8...
-
- "git commit c5953fd0a23d8a0dc9475dfa76ad0dfd07663717":https://github.com/illumos/illumos-gate/commit/c5953fd0a23d8a0d...
-
- "git commit 9a634533d15821efb93a491af59ea24d69227322":https://github.com/illumos/illumos-gate/commit/9a634533d15821ef...
-
- "git commit 9557befbc07a121f4baa233e966d24be5dca7345":https://github.com/illumos/illumos-gate/commit/9557befbc07a121f...
-
- "git commit 5539384561c0efe744226d5e5d331f5546e500e4":https://github.com/illumos/illumos-gate/commit/5539384561c0efe7...
2014-04-30
-
- sema_init(3C) contains a synopsis of sema_init that says:...
-
- When using mdb -k to poke at certain processes, I almost always do things like, '::pgrep -z test', only to then remem...
-
08:13 PM Bug #4827: nfs4: slow file locking
- The code was introduced by Git commit 7c478bd95313f5f23a4c, so I think the code is at least 9 years old. To start wit...
-
- With multiple concurrent processes, the nfs4 file locking gets unacceptable slow for us.
In case we cannot get the... -
03:51 PM Bug #4785: mdb crashes in ::ldi_handle
- I don't know if this is the correct fix (move mdb_getopts behind DCMD_ADDRSPEC) or if mdb_getopts should be modified....
-
-
- During ILB tests, it was found that ILB full NAT simply corrupts vmem arena causing random panic (more frequently obs...
2014-04-29
-
- "git commit 19449258028e6813f0b7a606b554b2fa37a390ec":https://github.com/illumos/illumos-gate/commit/19449258028e6813...
-
10:36 PM Bug #4825 (New): mpt_sas should not do IO completions in the ISR
- We have seen a performance problem caused by spending too much time in interrupt context when doing lots of writes on...
-
- "git commit cb3e7fb42f8104f779abb6856ccf6e5b8e6419d8":https://github.com/illumos/illumos-gate/commit/cb3e7fb42f8104f7...
-
- "git commit b59e2127f21675e88c58a4dd924bc55eeb83c7a6":https://github.com/illumos/illumos-gate/commit/b59e2127f21675e8...
-
- "git commit b59e2127f21675e88c58a4dd924bc55eeb83c7a6":https://github.com/illumos/illumos-gate/commit/b59e2127f21675e8...
-
- "git commit b59e2127f21675e88c58a4dd924bc55eeb83c7a6":https://github.com/illumos/illumos-gate/commit/b59e2127f21675e8...
-
- "git commit a833a696f1726fd5d95ded0820612f465a2dad8d":https://github.com/illumos/illumos-gate/commit/a833a696f1726fd5...
-
- "git commit 9b1d70f8223c4278acf4ca60eaf5dd285f72eeee":https://github.com/illumos/illumos-gate/commit/9b1d70f8223c4278...
-
- "git commit 6eedf6a58bb97d935b764a55719f82a486168fbb":https://github.com/illumos/illumos-gate/commit/6eedf6a58bb97d93...
-
- "git commit 6eedf6a58bb97d935b764a55719f82a486168fbb":https://github.com/illumos/illumos-gate/commit/6eedf6a58bb97d93...
-
-
-
-
-
-
2014-04-28
-
- Somehow (I'm not sure how), we have a priv_basicset(3C), which refers to a page which doesn't actually describe priv_...
-
- Use these macros instead of open-coding the division & multiplication.
-
- There are a couple of instances of nfs4 code doing:...
-
- We have filebench 1.4.8. The community has moved on and newer releases (1.4.9.1 has existed for a couple of years) a...
-
- That's pretty much why I filed the bug. I had a script that called 'rm -f' but it didn't have permission to remove t...
-
01:30 PM Bug #4816: rm does not print error messages when -f is used
- About all I can add is the typical use in a shell script. The construct:
@
if [ -f $F ]
then
rm $F
done
@
... -
11:59 AM Bug #4820 (New): Panic in zfs_ereport_start_checksum when checksum error is EINVAL
- Crash stack is the following:
ffffff005ca12cc0 zfs_ereport_start_checksum+0xa6(ffffff0d34a0f000, 0, ffffff11106134... -
- Maybe we don't want opendirat after all. The GNU libc developers have decided not to make it public.
-
- webrev: http://ma.nexenta.com/~woodstock/mptsas-timeout/
This includes the following commits from Nexentas nza-ke... -
- The existing timeout handling in mpt_sas are based on a per-target counter which is decremented every watchdog interv...
-
- Code reviews here:
http://cr.illumos.org/~webrev/gdamore/printf2/
and
https://bitbucket.org/gdamore/illumos... -
- This was pointed out by folks looking to integrate printf(1) into FreeBSD, who noted that the n$ support in my printf...
2014-04-27
-
-
- In the usr/src/lib/README.mapfiles there is this example:...
2014-04-26
-
- Yes, but it only says "in the case of non-existent operands". What about other cases such as permission errors?
-
- This exerpt from the man page seems clear to me:
@
/usr/xpg4/bin/rm
The following options are supported... -
- When the -f option is passed to rm it does not print error messages meaning the only way to spot an error is to check...
-
- I have encountered a number of scripts of both Debian and Gentoo that use GNU rm's -v option. This causing rm to pri...
-
-
-
- "git commit 6e987ca0b653b6724551e3e999d35f05512a5a6a":https://github.com/illumos/illumos-gate/commit/6e987ca0b653b672...
-
- DEBUG builds spit out the above on Thinkpad T520.
-
- After the nisplus removal in December 2009 (Sun bug 6874309) there were left some supporting functions in libnsl, for...
-
- "git commit 9c720e3bfb657054035cb780c4d222291b7d4917":https://github.com/illumos/illumos-gate/commit/9c720e3bfb657054...
-
- Because hrtime_t is defined as a *signed* 64-bit int:
http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/...
2014-04-25
-
11:59 PM Bug #4809: NANOSEC should be 'long long' to avoid integer overflow bugs
- If the expected result is most often an hrtime_t, how about using the ULL suffix?
-
- NANOSEC is defined as: 1000000000. This makes it an int - a 32-bit signed integer (on both ILP32 and LP64).
If NA... -
- tv2ns multiplies time_t with NANOSEC:...
-
- spa_async_tasks_pending multiplies NANOSEC by zfs_ccw_retry_interval (a global static int = 300), the compiler simply...
-
- In the usr/src/lib/libnsl/rpc/getdname.c file there is this piece of code:...
-
- Instead of issuing CPUID directly, apix/pcplusmp should use is_x86_feature().
-
- x2APIC CPUID flag should be decoded and presented via the x86_feature interface.
-
- The apic_mode global really should be an enum instead of just an int. This makes mdb's ::print output way easier to d...
-
- The apix and pcplusmp makefiles disable a number of warnings needlessly. Other warnings are pretty easy to address.
... -
- The return_xprt_copy() in libnsl locks svc_door_mutex and then calls svc_door_destroy():...
-
-
10:20 AM Bug #4722: fm/smtp-notify mails about events in previously closed cases upon fmd (re)start
- Jussi Sallinen wrote:
> When implementing this kind of behavior we must also be sure that the email is sent when the... -
03:32 AM Bug #4722: fm/smtp-notify mails about events in previously closed cases upon fmd (re)start
- When implementing this kind of behavior we must also be sure that the email is sent when the event is created & immed...
2014-04-24
-
- All the message appear to come from pcplusmp, not apix so I wouldn't expect tweaking apix to have any effect.
-
- In a case a strdup() will fail in _clnt_tli_create_timed() we will leak cl.
-
- the prototype for copyin(9F) seems to be in sys/systm.h. Now, perhaps I'm an idiot, but I did what the manual page t...
2014-04-23
-
-
- "git commit 6907ca4bc759c5fd81b8a60fc5e1a1c7d1972e6a":https://github.com/illumos/illumos-gate/commit/6907ca4bc759c5fd...
-
-
- "git commit 9aff5184bfbc5b708bf2ddd810ce726347b52332":https://github.com/illumos/illumos-gate/commit/9aff5184bfbc5b70...
-
-
- "git commit 4cde9a1e4370decb0ae4451cc85ce57730965c8c":https://github.com/illumos/illumos-gate/commit/4cde9a1e4370decb...
-
-
- "git commit 3b082ee621e893f087e4272c7e8b6d512d24e155":https://github.com/illumos/illumos-gate/commit/3b082ee621e893f0...
-
- http://cr.illumos.org/~webrev/risto3/il_4795/
-
12:48 PM Bug #4798 (New): Allow zfs filesystem_limit or snapshot_limit to be set on delegated dataset
- Currently the limits cannot be set for datasets that are under delegated datasets. We disallow this activity from wit...
-
- Add checks to zdb to verify that the fs/ss counts are correct. We visit all the dsl_dir's from dump_one_dir().
-
- "git commit a2afb611b30628fb74ad9eade4ae465f9031e262":https://github.com/illumos/illumos-gate/commit/a2afb611b30628fb...
-
2014-04-22
-
05:58 PM Bug #4767 (Closed): dtrace_probe() always has the timestamp
-
- "git commit 82d86f4e8b1400610e24f161032c16e049f173a0":https://github.com/illumos/illumos-gate/commit/82d86f4e8b140061...
-
- ...
-
-
- "git commit 136dc9cbe3ceda200434238d1015d39bca1d2d1a":https://github.com/illumos/illumos-gate/commit/136dc9cbe3ceda20...
-
-
- Duplicates #2897.
-
02:41 PM Bug #4794 (Rejected): zpool split undocumented
- @zpool@ shows a @split@ subcommand:...
-
-
-
-
01:27 AM Bug #4793 (New): spa_import_rootpool and vfs_mountroot rely on device path strings passed from GRUB
- A subject that I raised earlier in email, and apparently made no bug about it (can't find any): http://thr3ads.net/zf...
2014-04-21
-
- "git commit f4625c772597dc3b4c0331014447cb2120e17d43":https://github.com/illumos/illumos-gate/commit/f4625c772597dc3b...
-
- "git commit f4625c772597dc3b4c0331014447cb2120e17d43":https://github.com/illumos/illumos-gate/commit/f4625c772597dc3b...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- This is not a bug report but rather a question. I'm rejecting this bug.
-
- Agreed, this isn't a bug report but rather a question. Closing the bug.
-
- This bug concerns a closed driver. Since (1) we can't debug it, and (2) we have an open source alternative, I'm reje...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- This is an unused function that has several issues:
(1) it uses ddi_get_item(9f) to rate limit instead of gethrtim... -
03:00 PM Bug #4772 (Closed): $C prints no output on crash dump following dtrace panic()
- ...
-
- I don't think this issue makes sense after #3310 was resolved.
-
02:42 AM Bug #4518: lockd: Cannot establish NLM service over <file desc. 9, protocol udp>
- Marcel Telka wrote:
> Are you able to reproduce this issue?
I have this same issue.
@[ Apr 20 20:28:17 Executi...
2014-04-20
-
- ...
2014-04-19
2014-04-18
-
- webrev contains a bunch of code to handle usr/closed in a special way. This can all go away.
-
- ...
-
- "git commit b6240e830b871f59c22a3918aebb3b36c872edba":https://github.com/illumos/illumos-gate/commit/b6240e830b871f59...
-
- ...
-
06:14 AM Bug #4773: iSCSI as a boot device
- The same behaviour is in a real hardware (multi-core celeron)
network in VBox is configured as a bridge to igb inter...
2014-04-17
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- ddi_get_time() returns a Unix timestamp based on the current TOD. If the TOD setting changes for whatever reason (day...
-
- The manpage actively encourages developers to do something that's dangerous.
ddi_get_time() returns a Unix timesta... -
- By the way, if VirtualBox allows you to snapshot a running system, the snapshot may be a good replacement for a crash...
-
- I definitely concur that iSCSI boot is finicky. However, it is not as simple as mp vs. up config. I've successfully ...
-
- Cannot boot from an iSCSI target if the system (hardware) has more than one CPU,
single CPU systems can boot from it... -
- See my comment above. Without the changes make can't parse the Makefiles.
-
- Are the following changes really needed?...
-
- Webrev: http://cr.illumos.org/~webrev/andy_js/1681/
-
05:52 PM Bug #4775 (Closed): Typo in getaddrinfo(3SOCKET) man page
- In the return values section:
The order of the addresses returned i nthe fourth argument is discussed in the A... -
- tcp(7p) manpage has spelled timestamp as time stamp in several places over the manpage.
Replace the instances of tim... -
- Andrew Stormont wrote:
> I think you'd be better off trying to remove these java bits from the gate. There has alre... -
- "git commit 9278ddffeeeafc4766f74c6e2327b83ada62329a":https://github.com/illumos/illumos-gate/commit/9278ddffeeeafc47...
-
2014-04-16
-
05:47 PM Bug #4767: dtrace_probe() always has the timestamp
- When you use the "timestamp" variable in a dtrace program, dtrace_probe() records the timestamp (from dtrace_gethrtim...
-
- Matt Ahrens:...
-
- Sebastien Roy:...
-
-
- I think you'd be better off trying to remove these java bits from the gate. There has already been some talk of remo...
-
- During the patch tests two panic were detected on ILB-enabled zone shutdown:
Apr 15 10:13:18 illumos-dev0 genunix:... -
- Similar to the mdb dcmd sockparams.
-
- Please review update webrev:
http://mtc.md/sscdvp/webrevs/issue_4735/
Regards,
Serghei Samsi -
- System panic is detected after that patch applied:
panic message: BAD TRAP: type=e (#pf Page fault) rp=ffffff0006fb6... -
-
- There are basically two issues with handling of rfs4_start_time.
*Issue 1*
When the rfs4_start_time is set in r... -
- ...
-
- "git commit 1b497ab83e8f1c58bba5da59c649207a442a4720":https://github.com/illumos/illumos-gate/commit/1b497ab83e8f1c58...
-
- ...
2014-04-15
-
-
-
-
-
-
- test
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- Various virutalization systems need to be able to virutalize the tsc. As part of that, they have to handle the fact t...
-
- This is a duplicate of #4617.
-
- CONDVAR(9F) has no description of this function
-
- illumos was built with debug info and tried to boot in virtualbox
but it panics in uts/i86pc/os/mp_startup.c:707
as... -
- ...
-
- "git commit 7266c4fdd3d819c74f84745da19a4eb34bd75f2d":https://github.com/illumos/illumos-gate/commit/7266c4fdd3d819c7...
2014-04-14
-
- See above.
-
- "git commit f5b1cef2488dc579cb4312d49164f10debf8e97d":https://github.com/illumos/illumos-gate/commit/f5b1cef2488dc579...
-
05:26 PM Bug #4738 (Pending RTI): tape block number is set at wrong place
-
- Add a ::mac dee-command to print information from mac_impl_t entries in the mac_impl_cache.
-
- Matt Ahrens:...
-
- George Wilson:...
-
- George Wilson:...
-
- Original bug:...
-
- Matt Ahrens:...
-
- Analysis by Adam Leventhal:...
2014-04-13
-
- You can't put a static v6 address on an interface via ipadm, unless an addrconf address already exists. This isn't r...
2014-04-12
-
-
- I guess this is the closed source version. Added #3409 as related.
-
- ...
-
- Please review http://mtc.md/sscdvp/webrevs/issue_4749/
Regards,
Serghei Samsi
2014-04-11
-
- ILB daemon RSS continuously grows if it has active rules with health-check.
RSS growth during an hour:
102657 daemo... -
- There is already an enum that defines all the possible fault states for TOD. The global tod_faulted should use this ...
-
- These defines haven't been used by anything in a loooong time. Same goes for _[ds]size_limit externs.
-
- Some preprocessor definitions got left over after EXPORT_SRC got removed.
-
- There are a number of silly typos in the AVL code. Fix them.
-
-
- Closing. INS can be set. Not illumos problem.
-
- Closing. INS can be set. Not illumos problem.
-
- as part of 4719 update gate build environment to [open]jdk7
remove trace of default_java
-
- as part of RTI: 4719 update gate build environment to [open]jdk7
the deprecated /usr/j2se usage should be removed fr... -
- as part of 4719 update gate build environment to [open]jdk7
javadoc was noticed to have changed the generated files...
2014-04-10
-
- "git commit 0b5ce10aee80822ecc7df77df92a5e24078ba196":https://github.com/illumos/illumos-gate/commit/0b5ce10aee80822e...
-
06:36 PM Bug #4740: kstat(1M) prints wrong snaptime
- No problem. I haven't put together a patch yet because I found this while debugging another issue that I'm still wor...
-
- This bug was introduced by me. You plan to provide a patch in illumos-joyent or here? Thanks for the analysis.
-
- Here's a D script I've been using to trace kstat_read() invocations in a target process:...
-
- While working on 4735 I've observed ILB_RULE_HASH [which is used for looking up through ILB rules with VIP address as...
-
- Hey Alexander, I think that's probably fine. The only thing that'd change is our behaviour if the child process chan...
2014-04-09
-
- Currently tape block number is reset in a thread which waits for mover continue event. In theory if there are n threa...
-
- Yeah, I'd been wondering why we weren't doing that, and assuming there had to be some kind of reason. Perhaps there ...
-
- Possible working fix: ...
-
- in (the kernel) zone_enter, we do indeed explicitly join zsched's session and pg. Which sets us up for racing like t...
-
- So, this is zlogin.c's sig_forward actually passing the signal along. The question then becomes, "How the hell is in...
-
- a bit of playing with dtrace...
-
- bdha reported on twitter that holding ^C with zlogin with a shell loop will halt some percentage of his zones. This ...
-
08:14 PM Bug #4737 (New): dlmgtd process deadlocks.
- When performing the dlmgmt_zonehalt door command, a global write lock is held across a system call. The system call t...
-
- Some testing results are available:
Now ilb_probe with ICMP works as well as ping & traceroute -I commands:
roo... -
- Please review
http://mtc.md/sscdvp/webrevs/issue_4735/
Regards,
Serghei Samsi
-
- One of drawbacks is the stopping ilb_probe from work (for health-check configured with ICMP ping).
Also ping and tra... -
- Adding simple test program.
-
- Here is a cleaned up version of the changes which includes updated man pages: http://cr.illumos.org/~webrev/andy_js/4...
2014-04-08
-
- http://mtc.md/sscdvp/webrevs/issue_4726/
-
- Please review
http://mtc.md/sscdvp/webrevs/issue_4734/
Regards,
Serghei Samsi
-
- It mostly refers to ilbd' ilbd_scf_str_to_ip/ilbd_scf_ip_to_str.
As inet_pton(3xnet)/inet_ntop(3xnet) manual states,... -
- Please review:
http://mtc.md/sscdvp/webrevs/issue_4733/
Regards,
Serghei Samsi -
- Include different memory management fixes in ILB daemon (usr/src/cmd/cmd-inet/usr.lib/ilbd).
-
- I think illumos-userland is dead. Do we even need to keep the Sun dhcp server?
Also available in: Atom