kern/164256: zfs: device entry for volume is not created after zfs receive
>Number: 164256 >Category: kern >Synopsis: zfs: device entry for volume is not created after zfs receive >Confidential: no >Severity: non-critical >Priority: low >Responsible:freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Jan 17 19:50:09 UTC 2012 >Closed-Date: >Last-Modified: >Originator: Dmitry Afanasiev >Release:FreeBSD 9.0-STABLE amd64 >Organization: >Environment: FreeBSD green 9.0-STABLE FreeBSD 9.0-STABLE #82: Sat Jan 7 00:24:08 MSK 2012 root@green:/usr/obj/usr/src/sys/green amd64 >Description: Device entry for received from snapshot volume must be created, but not created. >How-To-Repeat: # zpool create tank `mdconfig -a -t swap -s 64m` # zfs create -V 1M tank/vol1 # zfs snapshot tank/vol1@snap # ls /dev/zvol/tank vol1vol1@snap # zfs send tank/vol1@snap | zfs receive tank/vol2 # ls /dev/zvol/tank vol1vol1@snap <- there is no entries for vol2 # zpool export tank # zpool import tank # ls /dev/zvol/tank vol1vol1@snap vol2vol2@snap <- magic! >Fix: >Release-Note: >Audit-Trail: >Unformatted: ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
misc/164370: zfs destroy for snapshot fails on i386
>Number: 164370 >Category: misc >Synopsis: zfs destroy for snapshot fails on i386 >Confidential: no >Severity: serious >Priority: medium >Responsible:freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Jan 22 11:30:12 UTC 2012 >Closed-Date: >Last-Modified: >Originator: Dmitry Afanasiev >Release:FreeBSD 9.0-STABLE i386 >Organization: >Environment: FreeBSD kot 9.0-STABLE FreeBSD 9.0-STABLE #55: Fri Jan 6 16:07:09 MSK 2012 root@kot:/usr/obj/usr/src/sys/kot i386 >Description: zfs destroy without -R options fails for any snapshot on _i386_ systems. I have no this problem on my amd64 and sparc64 systems with same source code level. >How-To-Repeat: # zpool create tank `mdconfig -a -t swap -s 64m` # zfs snapshot tank@snap # zfs destroy -v tank@snap will destroy tank@snap will reclaim 0 # echo $? 1 # zfs destroy -rv tank@snap will destroy tank@snap will reclaim 0 # echo $? 1 # zfs destroy -Rv tank@snap will destroy tank@snap will reclaim 0 # echo $? 0 >Fix: >Release-Note: >Audit-Trail: >Unformatted: ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
Re: misc/164370: zfs destroy for snapshot fails on i386
The following reply was made to PR misc/164370; it has been noted by GNATS. From: Dmitry Afanasiev To: bug-follo...@freebsd.org Cc: Subject: Re: misc/164370: zfs destroy for snapshot fails on i386 Date: Sun, 22 Jan 2012 17:30:20 +0400 Oops. I'm confused. Problem exists on sparc64 too. Only amd64 systems not affected by this issue. ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
kern/155227: netflow.c does not compile with disabled INET6
>Number: 155227
>Category: kern
>Synopsis: netflow.c does not compile with disabled INET6
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible:freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Thu Mar 03 14:50:09 UTC 2011
>Closed-Date:
>Last-Modified:
>Originator: Dmitry Afanasiev
>Release:9.0-CURRENT
>Organization:
>Environment:
FreeBSD orion 9.0-CURRENT FreeBSD 9.0-CURRENT #0: Wed Mar 2 12:54:04 MSK 2011
root@orion:/usr/obj/usr/src/sys/orion amd64
>Description:
sys/netgraph/netflow/netflow.c does not compile with disabled INET6:
mno-sse -mno-sse2 -mno-sse3 -msoft-float -fno-asynchronous-unwind-tables
-ffreestanding -std=iso9899:1999 -Wall -Wredundant-decls -Wnested-externs
-Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual
-Wundef -Wno-pointer-sign -fformat-extensions -c
/usr/src/sys/modules/netgraph/netflow/../../../netgraph/netflow/netflow.c
/usr/src/sys/modules/netgraph/netflow/../../../netgraph/netflow/netflow.c: In
function 'expire_flow':
/usr/src/sys/modules/netgraph/netflow/../../../netgraph/netflow/netflow.c:286:
error: 'struct netflow' has no member named 'zone6'
/usr/src/sys/modules/netgraph/netflow/../../../netgraph/netflow/netflow.c:301:
error: 'struct netflow' has no member named 'zone6'
*** Error code 1
1 error
*** Error code 2
1 error
*** Error code 2
1 error
*** Error code 2
1 error
*** Error code 2
1 error
*** Error code 2
1 error
>How-To-Repeat:
1. Update to r219182
2. Disable INET6
3. Build kernel
>Fix:
Patch attached with submission follows:
--- netflow.c.orig 2011-03-03 16:36:00.895086339 +0300
+++ netflow.c 2011-03-03 16:37:32.335079030 +0300
@@ -282,8 +282,10 @@
atomic_add_32(&priv->info.nfinfo_export9_failed, 1);
if (version == IPVERSION)
uma_zfree_arg(priv->zone, fle, priv);
+#ifdef INET6
else if (version == IP6VERSION)
uma_zfree_arg(priv->zone6, fle, priv);
+#endif
else
panic("ng_netflow: Unknown IP proto: %d",
version);
return;
@@ -297,8 +299,10 @@
if (version == IPVERSION)
uma_zfree_arg(priv->zone, fle, priv);
+#ifdef INET6
else if (version == IP6VERSION)
uma_zfree_arg(priv->zone6, fle, priv);
+#endif
}
/* Get a snapshot of node statistics */
>Release-Note:
>Audit-Trail:
>Unformatted:
___
freebsd-bugs@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
misc/156082: gcc makes incorrect code when swapping array elements by xor without optimization
>Number: 156082
>Category: misc
>Synopsis: gcc makes incorrect code when swapping array elements by xor
>without optimization
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible:freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Thu Mar 31 08:50:12 UTC 2011
>Closed-Date:
>Last-Modified:
>Originator: Dmitry Afanasiev
>Release:FreeBSD 9.0-CURRENT i386
>Organization:
>Environment:
FreeBSD orion 9.0-CURRENT FreeBSD 9.0-CURRENT #6: Sun Mar 27 03:24:55 MSD 2011
root@orion:/usr/obj/usr/src/sys/orion amd64
>Description:
Base gcc 4.2.1 makes incorrect code when swapping array elements by xor without
optimization.
Simple .c code attached as xor.txt.
Checked on FreeBSD 9.0 on i386/amd64/sparc64 platforms.
>How-To-Repeat:
# gcc -o xor xor.txt
# gcc -o xoro1 -O1 xor.txt
# ./xor
a0 = 127 b0 = 33
a1 = 127 b1 = 33
a2 = 127 b2[0] = 33
a3 = 127 b3[0] = 0
# ./xoro1
a0 = 127 b0 = 33
a1 = 127 b1 = 33
a2 = 127 b2[0] = 33
a3 = 127 b3[0] = 33
But in theory output must be identical.
>Fix:
Patch attached with submission follows:
#include
#include
int
main(int argc, char *argv[])
{
int a0 = 33;
int b0 = 127;
int a1 = 33;
int b1 = 127;
int a2 = 33;
int b2[] = {127};
int a3 = 33;
int b3[] = {127};
a0 ^= b0 ^= a0 ^= b0;
b1 ^= a1 ^= b1 ^= a1;
a2 ^= b2[0] ^= a2 ^= b2[0];
b3[0] ^= a3 ^= b3[0] ^= a3;
printf("a0 = %i b0 = %i\n", a0, b0);
printf("a1 = %i b1 = %i\n", a1, b1);
printf("a2 = %i b2[0] = %i\n", a2, b2[0]);
printf("a3 = %i b3[0] = %i\n", a3, b3[0]);
return EX_OK;
}
>Release-Note:
>Audit-Trail:
>Unformatted:
___
freebsd-bugs@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
bin/145234: zvol with org.freebsd:swap=on crashes zfs list
>Number: 145234 >Category: bin >Synopsis: zvol with org.freebsd:swap=on crashes zfs list >Confidential: no >Severity: serious >Priority: medium >Responsible:freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Wed Mar 31 09:00:14 UTC 2010 >Closed-Date: >Last-Modified: >Originator: Dmitry Afanasiev >Release:FreeBSD 9.0-CURRENT amd64 >Organization: >Environment: FreeBSD green 9.0-CURRENT FreeBSD 9.0-CURRENT #66: Tue Mar 30 19:33:40 MSD 2010 r...@green:/usr/obj/usr/src/sys/green amd64 >Description: Only on fresh current: # zfs create -V 128m rpool/test # zfs list rpool/test NAME USED AVAIL REFER MOUNTPOINT rpool/test 128M 55.9G16K - # zfs set org.freebsd:swap=on rpool/test # zfs list rpool/test Segmentation fault (core dumped) # zfs list Segmentation fault (core dumped) # zfs inherit org.freebsd:swap rpool/test # zfs list rpool/test NAME USED AVAIL REFER MOUNTPOINT rpool/test 128M 55.9G16K - backtrace looks like this: (gdb) bt #0 0x000800662730 in zfs_prune_proplist () from /lib/libzfs.so.2 #1 0x0040abf2 in ?? () #2 0x000800664694 in zfs_iter_filesystems () from /lib/libzfs.so.2 #3 0x0040abe1 in ?? () #4 0x00080065718c in zfs_iter_root () from /lib/libzfs.so.2 #5 0x0040ae7f in ?? () #6 0x00406a70 in ?? () #7 0x00408639 in ?? () #8 0x00403bfe in ?? () #9 0x00080053b000 in ?? () #10 0x in ?? () #11 0x in ?? () #12 0x0002 in ?? () #13 0x7fffeeb0 in ?? () #14 0x7fffeeb4 in ?? () #15 0x in ?? () #16 0x7fffeeb9 in ?? () .. I have this problem on fresh amd64 and sparc64 systems. zfs get all rpool/test works without problems. Also this fails automatic swap setup in /etc/rc.d/zfs. >How-To-Repeat: >Fix: >Release-Note: >Audit-Trail: >Unformatted: ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
Re: bin/145234: zvol with org.freebsd:swap=on crashes zfs list
The following reply was made to PR bin/145234; it has been noted by GNATS. From: Dmitry Afanasiev To: bug-follo...@freebsd.org Cc: Subject: Re: bin/145234: zvol with org.freebsd:swap=on crashes zfs list Date: Wed, 31 Mar 2010 14:16:53 +0400 I'm sorry, /etc/rc.d/zvol is a correct script name, that fails to configure swap area. ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
bin/173298: Splitted pool is not expandable
>Number: 173298 >Category: bin >Synopsis: Splitted pool is not expandable >Confidential: no >Severity: non-critical >Priority: low >Responsible:freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Fri Nov 02 19:10:00 UTC 2012 >Closed-Date: >Last-Modified: >Originator: Dmitry Afanasiev >Release:FreeBSD 9.0-STABLE >Organization: >Environment: FreeBSD green 9.0-STABLE FreeBSD 9.0-STABLE #87: Wed Jun 20 11:20:29 MSK 2012 root@green:/usr/obj/usr/src/sys/green amd64 >Description: No way to expand pool after zpool split >How-To-Repeat: # dd if=/dev/zero of=/var/tmp/tank bs=1024k count=64 # dd if=/dev/zero of=/var/tmp/tank2 bs=1024k count=128 # zpool create tank /var/tmp/tank # zpool list tank NAME SIZE ALLOC FREECAP DEDUP HEALTH ALTROOT tank 59.5M 89.5K 59.4M 0% 1.00x ONLINE - # zpool attach tank /var/tmp/tank /var/tmp/tank2 Wait for resilvering to be complete # zpool split -o autoexpand=on tank tank2 /var/tmp/tank2 # zpool import -d /var/tmp tank2 # zpool list tank2 NAMESIZE ALLOC FREECAP HEALTH ALTROOT tank2 59,5M 196K 59,3M 0% ONLINE - We have no way to expand pool tank2 Attach new device, set autoexpand property and then detach old device works properly: # zpool attach tank /var/tmp/tank /var/tmp/tank2 # zpool set autoexpand=on tank # zpool detach tank /var/tmp/tank # zpool list tank NAME SIZE ALLOC FREECAP DEDUP HEALTH ALTROOT tank 124M 133K 123M 0% 1.00x ONLINE - >Fix: >Release-Note: >Audit-Trail: >Unformatted: ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
Re: bin/173298: Splitted pool is not expandable
The following reply was made to PR bin/173298; it has been noted by GNATS. From: Dmitry Afanasiev To: Andriy Gapon Cc: bug-follo...@freebsd.org Subject: Re: bin/173298: Splitted pool is not expandable Date: Mon, 12 Nov 2012 12:04:46 +0400 On 03.11.2012 01:33, Andriy Gapon wrote: >> We have no way to expand pool tank2 > Have you tried zpool online -e ... ... ? Omg. I'm confused, solaris's zpool command does not have '-e' switch. Yes, zpool online -e expands pool, but... Why autoexpand property for spitted pool have no effect? ___ freebsd-bugs@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
