ANNOUNCE: nfs-utils-1.2.6 released.

DISCUSSION Steve Dickson 4 participants 9 messages May 14, 2012 – May 17, 2012

Subsystems (2)

NFS client NFS server

lore

Share & tools

lore

Steve Dickson

author

2012-05-14 19:16 UTC #0 lore raw

New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard
* The addition of -l option to rpcgssd to enforce legacy behaviour

As well as the usual assorted bug fixes...

The tarball can only be found at Sourceforge.net since still unable to access ftp://ftp.kernel.org (I'm working on it):
http://sourceforge.net/projects/nfs

The git tree is at:
git://linux-nfs.org/nfs-utils

The change log is also only on Sourceforge.net:
http://sourceforge.net/projects/nfs/files/nfs-utils/1.2.6/Changelog-1.2.6/download

steved.

NeilBrown

2012-05-15 04:06 UTC #1 lore raw

Quoted 13 lines · On Mon, 14 May 2012 15:15:38 -0400 Steve Dickson <[email protected]> wrote:

New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard
* The addition of -l option to rpcgssd to enforce legacy behaviour
As well as the usual assorted bug fixes...
The tarball can only be found at Sourceforge.net since still unable to access ftp://ftp.kernel.org (I'm working on it):
http://sourceforge.net/projects/nfs

Quoted 8 lines

The git tree is at:
git://linux-nfs.org/nfs-utils
The change log is also only on Sourceforge.net:
   http://sourceforge.net/projects/nfs/files/nfs-utils/1.2.6/Changelog-1.2.6/download
steved.

Thanks Steve.

I noticed that the "default download" was still 1.2.5.

I wondered why and eventually found this:

https://sourceforge.net/apps/trac/sourceforge/ticket/20991

So I took the liberty of making the appropriate change, and now the default download is 1.2.6.
I hope that's OK.

NeilBrown

Steve Dickson

author

2012-05-15 11:51 UTC #2 lore raw

Quoted 28 lines · NeilBrown

Quoted 13 lines · On Mon, 14 May 2012 15:15:38 -0400 Steve Dickson <[email protected]> wrote:
New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard
* The addition of -l option to rpcgssd to enforce legacy behaviour
As well as the usual assorted bug fixes...
The tarball can only be found at Sourceforge.net since still unable to access ftp://ftp.kernel.org (I'm working on it):
http://sourceforge.net/projects/nfs
Quoted 8 lines
The git tree is at:
git://linux-nfs.org/nfs-utils
The change log is also only on Sourceforge.net:
   http://sourceforge.net/projects/nfs/files/nfs-utils/1.2.6/Changelog-1.2.6/download
steved.
Thanks Steve.
I noticed that the "default download" was still 1.2.5.

Quoted 7 lines

I wondered why and eventually found this:
https://sourceforge.net/apps/trac/sourceforge/ticket/20991
So I took the liberty of making the appropriate change, and now the default download is 1.2.6.
I hope that's OK.

Definitely! I was assuming it would automatic change after a period of time... Thanks for taking care of it!

steved.

Jim Rees

2012-05-15 12:20 UTC #3 lore raw

In reply to Steve Dickson

If you're going to use the block layout driver, you'll also want to apply this patch on top of 1.2.6 to prevent your log from filling with garbage.

Quoted 1 line

From 76e9f814b0c6a9ef996b14123780bb39f448cafb Mon Sep 17 00:00:00 2001

From: Jim Rees <[email protected]>
Date: Tue, 27 Mar 2012 10:18:39 -0400
Subject: [PATCH 1/2] remove pretty_sig()

This really only works against EMC servers. There is at least one server that returns unprintable signatures, which fill the log with garbage (the spec does not mandate printable signatures). It could be made more generic, for example by checking each byte for isprint(). But the signatures are really only of interest to developers, not admins, so it seems better to just remove it.

Signed-off-by Jim Rees <[email protected]>

---
 utils/blkmapd/device-process.c |   29 -----------------------------
 1 file changed, 29 deletions(-)

utils/blkmapd/device-process.c +0 ~~-29~~

index 652a7a8..5fe3dff 100644
@@ -49,28 +49,6 @@
 
 #include "device-discovery.h"
 
-static char *pretty_sig(char *sig, uint32_t siglen)
-{
-	static char rs[100];
-	uint64_t sigval;
-	unsigned int i;
-
-	if (siglen <= sizeof(sigval)) {
-		sigval = 0;
-		for (i = 0; i < siglen; i++)
-			sigval |= ((unsigned char *)sig)[i] << (i * 8);
-		sprintf(rs, "0x%0llx", (unsigned long long) sigval);
-	} else {
-		if (siglen > sizeof rs - 4) {
-			siglen = sizeof rs - 4;
-			sprintf(&rs[siglen], "...");
-		} else
-			rs[siglen] = '\0';
-		memcpy(rs, sig, siglen);
-	}
-	return rs;
-}
-
 uint32_t *blk_overflow(uint32_t * p, uint32_t * end, size_t nbytes)
 {
 	uint32_t *q = p + ((nbytes + 3) >> 2);

@@ -109,9 +87,6 @@ static int decode_blk_signature(uint32_t **pp, uint32_t * end,
 		 * for mapping, then thrown away.
 		 */
 		comp->bs_string = (char *)p;
-		BL_LOG_INFO("%s: si_comps[%d]: bs_length %d, bs_string %s\n",
-			    __func__, i, siglen,
-			    pretty_sig(comp->bs_string, siglen));
 		p += ((siglen + 3) >> 2);
 	}
 	*pp = p;

@@ -152,10 +127,6 @@ read_cmp_blk_sig(struct bl_disk *disk, int fd, struct bl_sig_comp *comp)
 	}
 
 	ret = memcmp(sig, comp->bs_string, siglen);
-	if (!ret)
-		BL_LOG_INFO("%s: %s sig %s at %lld\n", __func__, dev_name,
-			    pretty_sig(sig, siglen),
-			    (long long)comp->bs_offset);
 
  out:
 	if (sig)

Signature · 1 line

1.7.9.5

Steve Dickson

author

2012-05-15 14:20 UTC #4 lore raw

Hey Jim,

Quoted 8 lines · Jim Rees

If you're going to use the block layout driver, you'll also want to apply this patch on top of 1.2.6 to prevent your log from filling with garbage.
From 76e9f814b0c6a9ef996b14123780bb39f448cafb Mon Sep 17 00:00:00 2001
From: Jim Rees <[email protected]>
Date: Tue, 27 Mar 2012 10:18:39 -0400
Subject: [PATCH 1/2] remove pretty_sig()

Quoted 8 lines

This really only works against EMC servers. There is at least one server that returns unprintable signatures, which fill the log with garbage (the spec does not mandate printable signatures). It could be made more generic, for example by checking each byte for isprint(). But the signatures are really only of interest to developers, not admins, so it seems better to just remove it.
Signed-off-by Jim Rees <[email protected]>

First let me apologise for dropping this.... I have on idea why it did not make it on my TODO list... A total space out on my part!

I just committed it (d68e4b5) and I will make sure its include in update releases in Fedora....

Again, sorry for spacing out...

steved.

Quoted 58 lines

---
 utils/blkmapd/device-process.c |   29 -----------------------------
 1 file changed, 29 deletions(-)

utils/blkmapd/device-process.c +0 ~~-29~~

index 652a7a8..5fe3dff 100644
@@ -49,28 +49,6 @@
 
 #include "device-discovery.h"
 
-static char *pretty_sig(char *sig, uint32_t siglen)
-{
-	static char rs[100];
-	uint64_t sigval;
-	unsigned int i;
-
-	if (siglen <= sizeof(sigval)) {
-		sigval = 0;
-		for (i = 0; i < siglen; i++)
-			sigval |= ((unsigned char *)sig)[i] << (i * 8);
-		sprintf(rs, "0x%0llx", (unsigned long long) sigval);
-	} else {
-		if (siglen > sizeof rs - 4) {
-			siglen = sizeof rs - 4;
-			sprintf(&rs[siglen], "...");
-		} else
-			rs[siglen] = '\0';
-		memcpy(rs, sig, siglen);
-	}
-	return rs;
-}
-
 uint32_t *blk_overflow(uint32_t * p, uint32_t * end, size_t nbytes)
 {
 	uint32_t *q = p + ((nbytes + 3) >> 2);

@@ -109,9 +87,6 @@ static int decode_blk_signature(uint32_t **pp, uint32_t * end,
 		 * for mapping, then thrown away.
 		 */
 		comp->bs_string = (char *)p;
-		BL_LOG_INFO("%s: si_comps[%d]: bs_length %d, bs_string %s\n",
-			    __func__, i, siglen,
-			    pretty_sig(comp->bs_string, siglen));
 		p += ((siglen + 3) >> 2);
 	}
 	*pp = p;

@@ -152,10 +127,6 @@ read_cmp_blk_sig(struct bl_disk *disk, int fd, struct bl_sig_comp *comp)
 	}
 
 	ret = memcmp(sig, comp->bs_string, siglen);
-	if (!ret)
-		BL_LOG_INFO("%s: %s sig %s at %lld\n", __func__, dev_name,
-			    pretty_sig(sig, siglen),
-			    (long long)comp->bs_offset);
 
  out:
 	if (sig)

NeilBrown

2012-05-17 06:44 UTC #5 lore raw

In reply to Steve Dickson

Quoted 4 lines · On Mon, 14 May 2012 15:15:38 -0400 Steve Dickson <[email protected]> wrote:

New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard

Don't you hate it when people don't bother to test all your -rc's, then report bugs against your final release? Me too.

mkdir /tmp/dest
make DESTDIR=/tmp/dest install

fails because it tries to install osd_login in /tmp/dest/sbin/osd_login without first creating /tmp/dest/sbin. All other required directories get created first.

This patch - stealing an idea from utils/mount - seem to fix it.

Thanks,
NeilBrown

commit 03bb227402ab023f4badb515022d49f82e01ff8d

Author: Neil Brown <[email protected]>
Date:   Thu May 17 16:40:40 2012 +1000

osd_login - ensure /sbin is created before installation.

If we use a more standard approach to describing the osd_login script, the automake infrastructure will create /sbin before attempting installation.
This is important for: make DESTDIR=/empty-dir install

Signed-off-by NeilBrown <[email protected]>

utils/osd_login/Makefile.am +4 -7

index adc493a..d17ffa7 100644
@@ -1,12 +1,9 @@
 ## Process this file with automake to produce Makefile.in
 
-OSD_LOGIN_FILES= osd_login
+# These binaries go in /sbin (not /usr/sbin), and that cannot be
+# overriden at config time.
+sbindir = /sbin
 
-EXTRA_DIST= $(OSD_LOGIN_FILES)
-
-all-local: $(OSD_LOGIN_FILES)
-
-install-data-hook:
-	$(INSTALL) --mode 755 osd_login $(DESTDIR)/sbin/osd_login
+sbin_SCRIPTS = osd_login
 
 MAINTAINERCLEANFILES = Makefile.in

Jim Rees

2012-05-17 11:56 UTC #6 lore raw

NeilBrown wrote:

Quoted 4 lines · On Mon, 14 May 2012 15:15:38 -0400 Steve Dickson <[email protected]> wrote:

New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard

Don't you hate it when people don't bother to test all your -rc's, then report bugs against your final release? Me too.

mkdir /tmp/dest
make DESTDIR=/tmp/dest install

fails because it tries to install osd_login in /tmp/dest/sbin/osd_login without first creating /tmp/dest/sbin. All other required directories get created first.

This patch - stealing an idea from utils/mount - seem to fix it.

Please see my comment in response to Mike Frysinger's similar patch.

Date: Thu, 17 May 2012 07:53:30 -0400
From: Jim Rees <[email protected]>
Subject: Re: [PATCH] osd_login: fix DESTDIR install
To: Mike Frysinger <[email protected]>
Cc: [email protected]

You misspelled "overridden." Yes, I know you just copied this line from mount/Makefile.am, but there is no sense compounding the error.

Also it might be worth mentioning in the comment why we have this restriction.

Steve Dickson

author

2012-05-17 12:25 UTC #7 lore raw

In reply to NeilBrown

Hey Neil,

Quoted 7 lines · NeilBrown

Quoted 4 lines · On Mon, 14 May 2012 15:15:38 -0400 Steve Dickson <[email protected]> wrote:
New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard

Quoted 5 lines

Don't you hate it when people don't bother to test all your -rc's, then report bugs against your final release? Me too.
mkdir /tmp/dest
make DESTDIR=/tmp/dest install

I'll add this test to that makerel script you gave me way back when... So hopefully I'll this type of breakage before I make a release!

Quoted 6 lines

fails because it tries to install osd_login in /tmp/dest/sbin/osd_login without first creating /tmp/dest/sbin. All other required directories get created first.
This patch - stealing an idea from utils/mount - seem to fix it.

Thanks for that patch... Its been applied with the correct spelling of overridden! ;-)

steved.

Quoted 39 lines

Thanks,
NeilBrown
commit 03bb227402ab023f4badb515022d49f82e01ff8d
Author: Neil Brown <[email protected]>
Date:   Thu May 17 16:40:40 2012 +1000
osd_login - ensure /sbin is created before installation.
If we use a more standard approach to describing the osd_login script, the automake infrastructure will create /sbin before attempting installation.
This is important for: make DESTDIR=/empty-dir install
Signed-off-by NeilBrown <[email protected]>
utils/osd_login/Makefile.am +4 -7
index adc493a..d17ffa7 100644
@@ -1,12 +1,9 @@
 ## Process this file with automake to produce Makefile.in
 
-OSD_LOGIN_FILES= osd_login
+# These binaries go in /sbin (not /usr/sbin), and that cannot be
+# overriden at config time.
+sbindir = /sbin
 
-EXTRA_DIST= $(OSD_LOGIN_FILES)
-
-all-local: $(OSD_LOGIN_FILES)
-
-install-data-hook:
-	$(INSTALL) --mode 755 osd_login $(DESTDIR)/sbin/osd_login
+sbin_SCRIPTS = osd_login
 
 MAINTAINERCLEANFILES = Makefile.in

Sachin Bhamare

2012-05-17 17:15 UTC #8 lore raw

In reply to NeilBrown

Quoted 47 lines · NeilBrown

Quoted 4 lines · On Mon, 14 May 2012 15:15:38 -0400 Steve Dickson <[email protected]> wrote:
New Features in this release:
* The addition the nfsdcld daemon
* The addition of the osd_login script which is mandated by the pnfs-objects standard
Don't you hate it when people don't bother to test all your -rc's, then report bugs against your final release? Me too.
mkdir /tmp/dest
make DESTDIR=/tmp/dest install
fails because it tries to install osd_login in /tmp/dest/sbin/osd_login without first creating /tmp/dest/sbin. All other required directories get created first.
This patch - stealing an idea from utils/mount - seem to fix it.
Thanks,
NeilBrown
commit 03bb227402ab023f4badb515022d49f82e01ff8d
Author: Neil Brown <[email protected]>
Date:   Thu May 17 16:40:40 2012 +1000
osd_login - ensure /sbin is created before installation.
If we use a more standard approach to describing the osd_login script, the automake infrastructure will create /sbin before attempting installation.
This is important for: make DESTDIR=/empty-dir install
Signed-off-by NeilBrown <[email protected]>
utils/osd_login/Makefile.am +3 -1
index adc493a..d17ffa7 100644
@@ -1,12 +1,9 @@
 ## Process this file with automake to produce Makefile.in
 
-OSD_LOGIN_FILES= osd_login
+# These binaries go in /sbin (not /usr/sbin), and that cannot be
+# overriden at config time.
+sbindir = /sbin

utils/osd_login/Makefile.am line 9

Quoted 10 lines

utils/osd_login/Makefile.am +1 -6

 
-EXTRA_DIST= $(OSD_LOGIN_FILES)
-
-all-local: $(OSD_LOGIN_FILES)
-
-install-data-hook:
-	$(INSTALL) --mode 755 osd_login $(DESTDIR)/sbin/osd_login
+sbin_SCRIPTS = osd_login
 
 MAINTAINERCLEANFILES = Makefile.in

ACK-by Sachin Bhamare <[email protected]>

Thanks for fixing this. My bad for not testing it with a non-empty DESTDIR.

- sachin

-----
Sachin Bhamare
Panasas Inc.