mmmmn

版权声明：转载时请以超链接形式标明文章原始出处和作者信息及本声明
http://mmmmn.blogbus.com/logs/2937932.html

1、Re: File names change to upper case in USB drive

-----------------------------------------------------------------------------

---

From: Robert Lawhead <news0000.5.unixguru@xxxxxxxxxxxxxxx>
Date: Wed, 07 Jun 2006 21:14:09 -0700

-----------------------------------------------------------------------------

---
Keith Thompson wrote:

Robert Lawhead <news0000.5.unixguru@xxxxxxxxxxxxxxx> writes:

googler wrote:

This problem is strange. I am using a San Disk Mini Cruzer USB drive
(512 MB) to copy some files from a Windows system to a Solaris 10
system. After copying the files in the USB drive from the Windows
system, when I plug it into the Solaris system, I see that many (but
not all) file and directory names that were actually in lowercase have
changed to uppercase. So the file aaa becomes AAA.

Any idea why this is happening? How to fix this?

Thanks.

"man pcfs". Perhaps you should mount using PCFS_MNT_FOLDCASE option.

That causes all file names to be mapped to lower case, which probably
isn't what's wanted.

Normally FAT filesystems are case-preserving; files retain the case
with which they were created, though you can't have two files in the
same directory whose names differ only in case, and you can refer to a
file "foo.txt", as "FOO.TXT, etc. Perhaps there's a bug in the
Solaris implementation of pcfs?

A good workaround is to copy a zip file or tarball to the USB drive,
rather than copying the files directly. That way the contained file
names are stored in the file itself rather than in the filesystem.

Ken,
The following excerpt from the solaris express pcfs (7) man
page seems (at least to me) to account for the observation made by
the OP. I agree PCFS_MNT_FOLDCASE may not do just what OP wants...
Had he asked how to preserve names in transfer, rather than how to
supress peculiar case wrap of file names already on the USB drive,
I would have offered the same advise as you.

Bob
File Systems pcfs(7FS)

When creating a file name, pcfs creates a short file name if
it fits the DOS short file name format, otherwise it
creates a long file name. This is because long file names
take more directory space. Because the root directory of a
pcfs file system is fixed size, long file names in the root
directory should be avoided if possible.

When displaying file names, pcfs shows them exactly as they
are on the media. This means that short names are displayed
as uppercase and long file names retain their case. Earlier
versions of pcfs folded all names to lowercase, which can be
forced with the PCFS_MNT_FOLDCASE mount option. All file
name searches within pcfs, however, are treated as if they
were uppercase, so readme.txt and ReAdMe.TxT refer to the
same file.
.

-----------------------------------------------------------------------------

---

References:
File names change to upper case in USB drive
From: googler
Re: File names change to upper case in USB drive
From: Robert Lawhead
Re: File names change to upper case in USB drive
From: Keith Thompson
Prev by Date: Re: File names change to upper case in USB drive
Next by Date: Re: max. size limit for mblk's
Previous by thread: Re: File names change to upper case in USB drive
Next by thread: Re: File names change to upper case in USB drive
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: PEZ MP3 Player vs. Solaris 9(SPARC)
... use with my Sun Blade 100 running Solaris 9 4/04 with the latest ...

enable
debug output for the pcfs filesystem, ... Trying to mount it as a pcfs

filesystem
... pcfs for fat media with 2048 bytes/sector. ... (comp.unix.solaris)
Re: File names change to upper case in USB drive
... to copy some files from a Windows system to a Solaris 10 ... After

copying
the files in the USB drive from the Windows ... system, when I plug it into

the Solaris system,
I see that many (but ... Normally FAT filesystems are case-preserving; ...

(comp.unix.solaris)
Re: File names change to upper case in USB drive
... googler wrote: ... to copy some files from a Windows system to a Solaris
10 ... After copying the files in the USB drive from the Windows ... system,

when
I plug it into the Solaris system, I see that many (but ...

(comp.unix.solaris)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(01)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
-----
Re: File names change to upper case in USB drive

-----------------------------------------------------------------------------

---

From: js@xxxxxxxxxxxxxxx (Joerg Schilling)
Date: 8 Jun 2006 20:59:47 GMT

-----------------------------------------------------------------------------

---
In article <lnejy0ibjn.fsf@xxxxxxxxxxxxxxx>,
Keith Thompson <kst-u@xxxxxxx> wrote:

Robert Lawhead <news0000.5.unixguru@xxxxxxxxxxxxxxx> writes:

googler wrote:

This problem is strange. I am using a San Disk Mini Cruzer USB drive
(512 MB) to copy some files from a Windows system to a Solaris 10
system. After copying the files in the USB drive from the Windows
system, when I plug it into the Solaris system, I see that many (but
not all) file and directory names that were actually in lowercase have
changed to uppercase. So the file aaa becomes AAA.

Any idea why this is happening? How to fix this?

Normally FAT filesystems are case-preserving; files retain the case
with which they were created, though you can't have two files in the
same directory whose names differ only in case, and you can refer to a
file "foo.txt", as "FOO.TXT, etc. Perhaps there's a bug in the
Solaris implementation of pcfs?

This is a Microsoft bug.

M$ ignores the case and even lists filenames that fit into 8.3 as
lowercase.

M$ uncorrectly creates FAT uppercase names for a lowercase filename
that fits into 8.3

If you use Solaris to create the filenames, Solaris correctly uses VFAT
names ;-)

--
EMail:joerg@xxxxxxxxxxxxxxxxxxxxxxxxxxx (home) Jörg Schilling D-13353 Berlin
js@xxxxxxxxxxxxxxx (uni)
schilling@xxxxxxxxxxxxxxxxxxx (work) Blog: http://schily.blogspot.com/
URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
.

-----------------------------------------------------------------------------

---

References:
File names change to upper case in USB drive
From: googler
Re: File names change to upper case in USB drive
From: Robert Lawhead
Re: File names change to upper case in USB drive
From: Keith Thompson
Prev by Date: Re: Sun changes patch policy
Next by Date: Solaris 9
Previous by thread: Re: File names change to upper case in USB drive
Next by thread: Re: Sol 10 x86 not recognizing NIC
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: A word of caution
... I post from the Windows system because I can do it remotely. ... No use

trying
to synchronize what I've read/posted on the Mac system with that of the

Windows system. ...
Use Remote Desktop ... I'm actually thinking of loading x86 Solaris onto my

laptop.
... (comp.sys.mac.advocacy)
File names change to upper case in USB drive
... I am using a San Disk Mini Cruzer USB drive ... to copy some files from a

Windows
system to a Solaris 10 ... system, when I plug it into the Solaris system, I

see that
many (but ... (comp.unix.solaris)
Re: File names change to upper case in USB drive
... to copy some files from a Windows system to a Solaris 10 ... After

copying
the files in the USB drive from the Windows ... system, when I plug it into

the Solaris system,
I see that many (but ... Normally FAT filesystems are case-preserving; ...

(comp.unix.solaris)
Re: File names change to upper case in USB drive
... googler wrote: ... to copy some files from a Windows system to a Solaris
10 ... After copying the files in the USB drive from the Windows ... system,

when
I plug it into the Solaris system, I see that many (but ...

(comp.unix.solaris)
Possible to burn CD that is readable on both Windows & Solaris?
... I would like to burn some data CDs that will be readable on both Windows

...
The files have long filenames. ... including the long filenames, but it won't

mount on
Solaris, either ... (comp.unix.solaris)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(06)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      

2、Re: High TCP connect timeout rate questions

-----------------------------------------------------------------------------

---

From: Sam N <sun@xxxxxxxxxxxxxx>
Date: Thu, 08 Jun 2006 07:23:09 GMT

-----------------------------------------------------------------------------

---
Kyle Tucker wrote:
I have received this error on a port listened to by a Squid server. Is this

warning issued prior to reaching the point
where pending TCP requests would have been dropped? And if
that point is reached, would the counter for tcpListenDropQ0 shown in netstat

be greater than 0? Mine is at 0. Thanks.

Hi Kyle

It sounds like maybe the box is getting hammered - Is it really busy? There

are steps you can take to tune the stack on Solaris boxes, you can make

changes to set numerous variables in order to increase throughput.

Below is a script I use when setting up web servers (Sol 10) which seems to

increase throughput - YMMV of course, and I wouldn't blindly apply the

script, take some time to work through it and identify if it would be useful

to you. Some features may not work on older Solaris versions, again YMMV.

http://docs.sun.com/app/docs?q=tunable+parameters+solaris is very useful.

#!/bin/sh
#
# S70stacktune
# v 1.22 25/2/2006 SamN
# Tune TCP stack on busy Solaris 10 boxes.

fill () {
awk '{leninput=length($($NF)); fill=63-leninput ; for (i=1; i< fill; i++)

fillchar=fillchar"." ; printf $($NF) fillchar}'
}

setparams () {
VAL=`/usr/sbin/ndd -set $1 $2 $3`
printf "Value of $1 $2 is: " | fill
printf " ${VAL} ($3)\n"
}

# Set the tcp time wait interval
printf "Set the tcp time wait interval\n"
setparams /dev/tcp tcp_time_wait_interval 60000

# Speed up the flushing of half-closed connection in state FIN_WAIT_2
printf "Flushing of half-closed connection in state FIN_WAIT_2\n"
setparams /dev/tcp tcp_fin_wait_2_flush_interval 67500ms

# Set the receive and transmit window sizes
printf "Set the receive and transmit window sizes\n"
setparams /dev/tcp tcp_xmit_hiwat 400000
setparams /dev/tcp tcp_recv_hiwat 400000

# Set number of half-open connections
printf "Set number of half-open connections\n"
setparams /dev/tcp tcp_conn_req_max_q0 102400

# Set number of simultaneous connections
printf "Set number of simultaneous connections\n"
setparams /dev/tcp tcp_conn_req_max_q 102400

# Set the maximum buffer size
printf "Set the maximum buffer size\n"
setparams /dev/tcp tcp_max_buf 4194304

# Set the tcp time wait interval
printf "Set the tcp time wait interval\n"
setparams /dev/tcp tcp_time_wait_interval 60000

# Set TCP connection abort interval
printf "Set TCP connection abort interval\n"
setparams /dev/tcp tcp_ip_abort_interval 60000

# Set congestion window size
printf "Set congestion window size\n"
setparams /dev/tcp tcp_cwnd_max 2097152

# Set TCP Keepalive Interval
printf "Set TCP Keepalive Interval\n"
setparams /dev/tcp tcp_keepalive_interval 60000

printf "Ok, tuning complete\n"
.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: High TCP connect timeout rate questions
From: Kyle Tucker
References:
High TCP connect timeout rate questions
From: Kyle Tucker
Prev by Date: Re: select nic interface for outgoing IP messages
Next by Date: Re: Netstat Problem on solaris 5.8
Previous by thread: High TCP connect timeout rate questions
Next by thread: Re: High TCP connect timeout rate questions
Index(es):
Date
Thread
We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(05)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
---------
Re: High TCP connect timeout rate questions

-----------------------------------------------------------------------------

---

From: Sam N <sun@xxxxxxxxxxxxxx>
Date: Fri, 09 Jun 2006 06:46:24 GMT

-----------------------------------------------------------------------------

---
Kyle Tucker wrote:
Sam N <sun@xxxxxxxxxxxxxx> wrote:
Kyle Tucker wrote:
I have received this error on a port listened to by a Squid server. Is this

warning issued prior to reaching the point
where pending TCP requests would have been dropped? And if
that point is reached, would the counter for tcpListenDropQ0 shown in netstat

be greater than 0? Mine is at 0. Thanks.

It sounds like maybe the box is getting hammered - Is it really busy? There

are steps you can take to tune the stack on Solaris boxes, you can make

changes to set numerous variables in order to increase throughput.

I don't know how busy that particular box was at the time (no sar) and
it's at a customer site, but will watch it more closely in the future.

Below is a script I use when setting up web servers (Sol 10) which seems

[ script snipped ]

This looks very helpful.

http://docs.sun.com/app/docs?q=tunable+parameters+solaris is very useful.

Yes, I read a good amount on docs.sun.com and Google groups when this
occured. But I can't find a definitive answer to what actually occurs,
if anything, when the error is issued and what increments the above
counters in netstat. Any idea?

Thanks.

As far as I'm aware the alert is generated when the box thinks it's under a

SYN flood attack (and the unestablished connection queue becomes full).

/dev/tcp tcp_conn_req_max_q0 - unestablished connection queue (incomplete

handshake)
/dev/tcp tcp_conn_req_max_q - established connection queue

Are two parameters that you can tune in order to increase the queue sizes.

http://www.informit.com/articles/article.asp?p=101138&seqNum=5&rl=1 has more

info - it seems to be a pretty good article.

cheers

Sam

.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: High TCP connect timeout rate questions
From: Kyle Tucker
References:
High TCP connect timeout rate questions
From: Kyle Tucker
Re: High TCP connect timeout rate questions
From: Sam N
Re: High TCP connect timeout rate questions
From: Kyle Tucker
Prev by Date: Re: shared library in 32 bit and 64 bit
Next by Date: Re: t2000, again (vs. V440s?)
Previous by thread: Re: High TCP connect timeout rate questions
Next by thread: Re: High TCP connect timeout rate questions
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: High TCP connect timeout rate questions
... Is this warning issued prior to reaching the point ... where pending TCP
requests would have been dropped? ... shown in netstat be greater than 0? ...
make changes to set numerous variables in order to increase throughput. ...

(comp.unix.solaris)
Re: High TCP connect timeout rate questions
... Is this warning issued prior to reaching the point ... where pending
TCP requests would have been dropped? ... shown in netstat be greater than 0?

...
(comp.unix.solaris)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(08)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      

---------
Re: High TCP connect timeout rate questions

-----------------------------------------------------------------------------

---

From: kylet@xxxxxxxxxxxxx (Kyle Tucker)
Date: Fri, 9 Jun 2006 12:22:34 +0000 (UTC)

-----------------------------------------------------------------------------

---
In article <4P8ig.14582$qD.954@xxxxxxxxxxxxxxxxxxxx>,
Sam N <sun@xxxxxxxxxxxxxx> writes:

Sam N <sun@xxxxxxxxxxxxxx> wrote:

Kyle Tucker wrote:

I have received this error on a port listened to by a Squid
server. Is this warning issued prior to reaching the point
where pending TCP requests would have been dropped? And if
that point is reached, would the counter for tcpListenDropQ0
shown in netstat be greater than 0? Mine is at 0. Thanks.

It sounds like maybe the box is getting hammered - Is it really busy?
There are steps you can take to tune the stack on Solaris boxes, you can
make changes to set numerous variables in order to increase throughput.

I don't know how busy that particular box was at the time (no sar) and
it's at a customer site, but will watch it more closely in the future.

Below is a script I use when setting up web servers (Sol 10) which seems

[ script snipped ]

This looks very helpful.

http://docs.sun.com/app/docs?q=tunable+parameters+solaris is very useful.

Yes, I read a good amount on docs.sun.com and Google groups when this
occured. But I can't find a definitive answer to what actually occurs,
if anything, when the error is issued and what increments the above
counters in netstat. Any idea?

As far as I'm aware the alert is generated when the box thinks it's
under a SYN flood attack (and the unestablished connection queue becomes
full).

Right, and then I'd have thought netstat would have showed
tcpListenDropQ0 > 0. The box had not been rebooted between
the message and when I ran netstat. Maybe I'll set up a test
box with a very low setting and see if I can get the error
generated and get tcpListenDropQ0 incremented. Thanks.

--
- Kyle
.

-----------------------------------------------------------------------------

---

References:
High TCP connect timeout rate questions
From: Kyle Tucker
Re: High TCP connect timeout rate questions
From: Sam N
Re: High TCP connect timeout rate questions
From: Kyle Tucker
Re: High TCP connect timeout rate questions
From: Sam N
Prev by Date: Re: Patching strategy for Solaris 10 Core Install
Next by Date: Re: Patching strategy for Solaris 10 Core Install
Previous by thread: Re: High TCP connect timeout rate questions
Next by thread: where is the "unsetenv" c function
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: High TCP connect timeout rate questions
... Is this warning issued prior to reaching the point ... where pending
TCP requests would have been dropped? ... shown in netstat be greater than 0?

...
(comp.unix.solaris)
Re: High TCP connect timeout rate questions
... Is this warning issued prior to reaching the point ... would the counter

for
tcpListenDropQ0 shown in netstat be greater than 0? ... There are steps you

can take to
tune the stack on Solaris boxes, you can make changes to set numerous

variables in order to increase
throughput. ... As far as I'm aware the alert is generated when the box

thinks it's under a SYN
flood attack (and the unestablished connection queue becomes full). ...

(comp.unix.solaris)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(06)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      

3、Re: ufsdump very slow

-----------------------------------------------------------------------------

---

From: Dan Foster <usenet@xxxxxxxxxxx>
Date: Tue, 06 Jun 2006 15:49:54 -0500

-----------------------------------------------------------------------------

---
In article <4em6i7F1elij4U1@xxxxxxxxxxxxxx>, Michael Laajanen

<michael_laajanen@xxxxxxxxx> wrote:

I have noticed for some time(after changing tape device) that the new
tape (Ultrium 400GB compressed) is very little loaded(13%) during
ufsdump and that the disk that is beeing dumped is fully loaded(92%) but
with a very low transfer rate(4048), why is that so see below from
iostat during a level 0 dump.

The machine is a E250 and all disks in the machine are brand new Maxtor
10k 147GB disks exept for the root disk which is a original 26GB Seagate
10k drive.

extended device statistics
r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device
0.0 127.0 0.0 4064.1 0.0 0.1 0.0 1.1 0 13 rmt/0

4064.1 KB/sec divided by 127.0 writes/sec = 32 KB writes.

This is too small for LTO. LTO minimum for reasonable performance is 256
KB writes.

Performance is crawling because LTO drive has to slow down. You are not
feeding data to it fast enough for it to run at full speed.

Bottleneck here is how you are using ufsdump. Your tape drive and hard
drives are fast. :)

You might want to try ufsdump's 'b 512' argument to tell it to write
data to tape in 512 block chunks (256 KB).

ufsdump may limit it to 126 block chunks; not sure, so test and verify.

-Dan
.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: ufsdump very slow
From: Michael Laajanen
References:
ufsdump very slow
From: Michael Laajanen
Prev by Date: Re: gethostbyname broken under Solaris 8
Next by Date: Re: gethostbyname broken under Solaris 8
Previous by thread: ufsdump very slow
Next by thread: Re: ufsdump very slow
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
RE: 306GB drives!
... 72GB scsi sca drives are avaialble for about $235 which works out to

about ...
>> storage disk, such that the signature can be read by a signal detector.

... >>
LTO tape drives actively doing the restore? ... (comp.os.vms)
Re: Need help on Tape,Optical disk...
... I do have extensive experience with both disk and tape in many ... Disk

drives,
on the other hand, ... They're sending 5 MB/s to a 60 MB/s drive, ...

(comp.arch.storage)
Re: W9xNT MCA Adapters - SPOCK-206
... IDs and find out which one is disk, CD, tape, scanner and so on, to be

... don't
recall offhand if CD-ROM drives trigger the lights, ... I seldom look at the

disk light
either. ... (comp.sys.ibm.ps2.hardware)
TAPE OR DISK STORAGE? MAYBE BOTH
... Tape or disk storage? ... The choice of enterprise storage systems ...
disk systems might suggest. ... drives to today's DLT and SuperDLT ...

(sci.chem)
TAPE OR DISK STORAGE? MAYBE BOTH
... Tape or disk storage? ... The choice of enterprise storage systems ...
disk systems might suggest. ... drives to today's DLT and SuperDLT ...

(sci.physics)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(01)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
-----
Re: ufsdump very slow

-----------------------------------------------------------------------------

---

From: jayl-news@xxxxxxxxxxxxxx
Date: 9 Jun 2006 14:56:12 -0700

-----------------------------------------------------------------------------

---
Michael Laajanen wrote:

I don't know what is average, it around 480000 files which uses 35GB of
data.

Ummmm, well then, yes you *do* know the average, it's 73KB,
and yes, that qualifies as lots of little files. There is nothing
that will back this up as fast as you (apparently) want. Star
will be faster than gtar or ufsdump, but not 10X faster or
anything like that.

Search this group for "Backup tools too slow" and you'll get some
education.

service in spare time in our company, otherwise we would have MS
everywhere :)

But the backups would still be slow. :-)

2. If installa 5'th disk in the E250(which makes a total of 5 147 GB
disks), would some kind of SW raid help this?

Striping would help a little bit, but not 1.5X or anything like that.
It doesn't sound like you're ready for that in any case.

3. We where planning to more to S10 in december last year, but S10 was
delay(update) so we are still on S9, would S10 and ZFS help us?

You're getting 4MB/s IIRC, that's not bad. ZFS might help a little,
but ZFS provides flexibility and control, not so much improved
speed. ZFS cannot change average seek time on your drives.

What you *can* do is reduce the amount of time that live
data changes on the drive reduce the integrity of your
backup. The "backup window". You do this by either taking
a snapshot of the file system, and backing up the snapshot,
or doing an additional mirror of the file system, and do a
detach/backup/re-attach of that extra mirror.

ZFS makes the snapshot pretty easy. You can snapshot
your UFS filesystems, too, see fssnap. Last time I tried
incremental ufsdump on fssnap snapshots, it didn't handle
the timestamps correctly, but that was on Sol8, maybe
it's fixed now.

I reiterate, you should search this group, you'll get a lot
of good information.

-Jay-

.

-----------------------------------------------------------------------------

---

References:
ufsdump very slow
From: Michael Laajanen
Re: ufsdump very slow
From: Dan Foster
Re: ufsdump very slow
From: Michael Laajanen
Re: ufsdump very slow
From: Dan Foster
Re: ufsdump very slow
From: Michael Laajanen
Re: ufsdump very slow
From: Dan Foster
Re: ufsdump very slow
From: Michael Laajanen
Prev by Date: Re: How can I script shadow file modifications?
Next by Date: Re: How can I script shadow file modifications?
Previous by thread: Re: ufsdump very slow
Next by thread: Xorg on Solaris 8 SPARC
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: Volume Shadow Copy on XP/2K3
... Most modern 3rd party backup software are VSS requestors - like, say,

Veritas
... which involve the data storage of aforementioned database software. ...

This
means that the snapshot can possibly contain non-integral in-transition ...

(comp.arch.storage)
Re: ZFS: filesystem approach
... >>with their own ZFS filesystem, to see how usable it would actually be?

...
so that if you wanted to use a traditional backup ... and the snapshots

appear in
a special ... (comp.unix.solaris)
Re: Looking for advice on rsync
... > they were asking about was using rsync or some other mechanism to ... I
explained that the backup can't really be ... never even notice the snapshot

...
We can mount the snapshot as another filesystem, ... (comp.os.linux.misc)
Re: batch file to delete files based on criteria
... >> I need to throw together a simple batch file to backup a file. ...
>> schedule this via Task Scheduler to run every hour or so. ... > Create a

snapshot
backup when a file changes. ... >!Snapshot script from inside Windows

Explorer.
... (microsoft.public.win2000.cmdprompt.admin)
Re: What are the advantages of LVM?
... enough that the backup is not self-consistant. ... now you can backup
the snapshot ... where you might have multiple virtual Linux machines running

(or
not ... (Ubuntu)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(04)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      

4、Re: Networking, Zones, & Firewall Question w/ Solaris 10+

-----------------------------------------------------------------------------

---

From: aryzhov@xxxxxxxxx
Date: 6 Jun 2006 02:28:21 -0700

-----------------------------------------------------------------------------

---
Lion-O wrote:

It might be arguable that it's in fact *global* zone that routes the traffic
in this scenario. But look, global zone may have no valuable IP address

Then routing would be hardly possible since you /need/ an entry in your

routing
table to allow the global zone to access the non-global zone. So much for the
extra security...

I don't need an entry in routing table,
neither need I an IP address in the global zone,
in order to filter or RDR the packets for a non-global zone's subnet.

And when it comes to using routing daemons (in.routed / routed)
which can IMO be a very important part of a somewhat more advanced
router then your argument becomes plain out wrong.

Again, routing and firewalling are two very different activities.
Your statement is perfectly correct for routing but wrong
for packet filtering and NAT/RDR. Please give an example of
advanced firewall that needs a routing daemon.

There is no 'security advantage' to be gained here since the actual routing
will always be done on the global zone.

There's no routing in the global zone, but RDR tables only.
No routing daemons, either. Just a kernel module.
Totally different level in OS as well as in the IP stack.

The only real security advantage is to place network services
(http, mta, etc.) servers in your non-global zones.

I agree, but this has nothing to do with the topic.

Security advantage here is that administrator can avoid any
management traffic on subnets *and* any rule managenent activity
in virtual hosts being filtered, NATed or RDR'd.

Even though I use a separate Solaris 10 box as dedicated firewall,
I still consider it reasonable to configure the extranet and intranet
IP addresses in two separate non-global zones, and use a third
non-global zone (on a separate physical segment) for interactive
access.

From this management zone, I can ssh to a global zone via internal

virtual interface not visible from "production" zones.

Comparing with a traditional 3-segment firewall (intra, extra, mgmt),
I assign this 3 "functions" to 3 isolated machines, which gives me
some additional piece of mind.

Regards,
Andrei

.

-----------------------------------------------------------------------------

---

Prev by Date: Re: Faster backups - possibly Solaris pipe buffer size
Next by Date: Re: Size difference when copying a large file in solaris
Previous by thread: Solaris10 HW 06/06
Next by thread: gethostbyname broken under Solaris 8
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: Networking, Zones, & Firewall Question w/ Solaris 10+
... Then routing would be hardly possible since you /need/ an entry in your

routing ...
table to allow the global zone to access the non-global zone. ... There is no

'security
advantage' to be gained here since the actual routing ... Because the

non-global zone
is in the routing table of the global zone. ... (comp.unix.solaris)
Re: Zones within a Zone?
... non-global zone can't influence the global zone (hint: ...

(comp.unix.solaris)
Zones and SMF issue
... I've got a package I produced. ... It works great in the global zone.
... So pkgadd running in the non-global zone cannot add that file. ...

(comp.unix.solaris)
Re: Configuring packages from zone to zone
... >samba in the global zone, ... >to my non-global zone by default,
but when I went to edit my smb.conf ... Why did you install the sunfreeware

version
and not the version ... (comp.unix.solaris)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(01)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      

5、Re: how to find the total number of open file descriptors per application

-----------------------------------------------------------------------------

---

From: Robert Milkowski <rmilkowski-NO-SPAM@xxxxxxxx>
Date: Tue, 6 Jun 2006 04:30:59 +0000 (UTC)

-----------------------------------------------------------------------------

---
Rob <user@xxxxxxxx> wrote:

Hi,

How do I find the total number of file descriptors by application (just the
number not a list of open fd's)? I'm trying to monitor a server to see if
and when the rlims [in /etc/system (rlim_fd_max and rlim_fd_cur)] are
reached so that I can tune it.

This is Solaris 8 and it has lsof installed.

ls -l /proc/pid/fd/ | wc -l

however these wont't be only fds to files (network sockets too, etc).
Also 'man pfiles'.

--
Robert Milkowski
rmilkowskiADAADS@xxxxxxxx
http://milek.blogspot.com
.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: how to find the total number of open file descriptors per application
From: Rob
References:
how to find the total number of open file descriptors per application
From: Rob
Prev by Date: how to find the total number of open file descriptors per

application
Next by Date: Re: how to find the total number of open file descriptors per

application
Previous by thread: how to find the total number of open file descriptors per

application
Next by thread: Re: how to find the total number of open file descriptors per

application
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: file descriptor to a command
... Rob wrote: ... > I am trying to learn file descriptors and thought
it might possibly ... > be useful to use them as a tmp file. ... If you want
to do something like this, you should use "named pipes". ...

(comp.unix.shell)
Re: Finding No. of open file descriptors in a process
... Note that in Solaris 10 pfiles includes the file name: ... carnot% pfiles
`pgrep vi` ... Current rlimit: 256 file descriptors ... (comp.unix.solaris)
Re: Finding Process ID of Runaway files
... gerg@panix.com (Greg Andrews) writes: ... output of pfiles on Solaris
7 and Solaris 8 look the same to me: ... Current rlimit: 256 file descriptors

...
(comp.sys.sun.admin)
Re: Finding Process ID of Runaway files
... gerg@panix.com (Greg Andrews) writes: ... output of pfiles on Solaris
7 and Solaris 8 look the same to me: ... Current rlimit: 256 file descriptors

...
(comp.unix.programmer)
Re: Finding Process ID of Runaway files
... gerg@panix.com (Greg Andrews) writes: ... output of pfiles on Solaris
7 and Solaris 8 look the same to me: ... Current rlimit: 256 file descriptors

...
(comp.unix.solaris)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(04)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
-----

6、Re: New 1.4T SVM UFS fs does not have enough inodes

-----------------------------------------------------------------------------

---

From: Jim Prescott <jgp@xxxxxxxxxxxxxxxxxx>
Date: 5 Jun 2006 18:56:58 -0400

-----------------------------------------------------------------------------

---
In article <1149515332.749153.145950@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Dougii <DouglasDenny@xxxxxxxxx> wrote:

How do I get more inodes in a multi-terabyte file UFS system?

Ask Sun to remove the arbitrary minimum of "nbpi=1048576" for EFI
filesystems. From what I've heard the limit is there because large
filesystems with lots of inodes take a long time to fsck. To me
a potential performance issue might justify a warning message and
a change in defaults, but not a hard limit.

If you can get by with .99T you can get as many inodes as you need.
Based on the df output you presented your current inode density is
around 211k so you could use "newfs -i 100000" on the .99T filesystem
to get a little extra usable space by making fewer inodes.

ZFS doesn't suffer from this limitation. I've also heard that with
the sources from OpenSolaris it is easy to bypass the EFI inode limit.
--
Jim Prescott - Computing and Networking Group jgp@xxxxxxxxxxxxxxxxxx
School of Engineering and Applied Sciences, University of Rochester, NY
.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Peter Eriksson
References:
New 1.4T SVM UFS fs does not have enough inodes
From: Dougii
Prev by Date: Re: gdb for Solaris 10
Next by Date: Re: Faster backups - possibly Solaris pipe buffer size
Previous by thread: New 1.4T SVM UFS fs does not have enough inodes
Next by thread: Re: New 1.4T SVM UFS fs does not have enough inodes
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: performance of filesystem xattrs with Samba4
... filesystems at: ... Andreas to add support to ext3 for large inodes.
... and only lost a very small amount when xattrs were not ... clearly how

much
we are paying for journaling. ... (Linux-Kernel)
Re: 1 Gigabyte SD card "full" after 467 MB [SOLVED]
... filesystems, including ext3, where it's possible to run out of inodes ...

In
practice most filesystems have huge numbers of inodes, ... the real problem

must have been
inodes, directory entries, etc. ... just have nautilus (or konqueror or any

other
application you want to ... (Ubuntu)
Re: All filesystems change to read-only after a while with kernel 2.6.6-1.435
... > of my filesystems (ext3) change to read-only. ... > about inodes
being members of orphan nodes. ... > that and the subsequent reboot works

fine. ...
(Fedora)
Re: New 1.4T SVM UFS fs does not have enough inodes
... Dougii wrote: ... filesystems with lots of inodes take a long time to
fsck. ... to get a little extra usable space by making fewer inodes. ... One

problem
with this many inodes is that fsck will use a *LOT* of RAM ...

(comp.unix.solaris)
Re: Can I use a round bobbin redirects in a shell ?
... names do not have permissions. ... > A name (struct dirent) is a pointer
to an inode. ... "File serial number", nothing about inodes. ... >> Where
is it stated in POSIX that all filesystems must use inodes? ...

(uk.comp.os.linux)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(02)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
----
Re: New 1.4T SVM UFS fs does not have enough inodes

-----------------------------------------------------------------------------

---

From: Peter Eriksson <peter@xxxxxxxxxx>
Date: Tue, 6 Jun 2006 05:17:55 +0000 (UTC)

-----------------------------------------------------------------------------

---
Jim Prescott <jgp@xxxxxxxxxxxxxxxxxx> writes:

In article <1149515332.749153.145950@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Dougii <DouglasDenny@xxxxxxxxx> wrote:

How do I get more inodes in a multi-terabyte file UFS system?

Ask Sun to remove the arbitrary minimum of "nbpi=1048576" for EFI
filesystems. From what I've heard the limit is there because large
filesystems with lots of inodes take a long time to fsck. To me
a potential performance issue might justify a warning message and
a change in defaults, but not a hard limit.

If you can get by with .99T you can get as many inodes as you need.
Based on the df output you presented your current inode density is
around 211k so you could use "newfs -i 100000" on the .99T filesystem
to get a little extra usable space by making fewer inodes.

ZFS doesn't suffer from this limitation. I've also heard that with
the sources from OpenSolaris it is easy to bypass the EFI inode limit.

Yup. It's really simple. You can get a premodified version from:

ftp://ftp.ifm.liu.se/pub/unix/mkfs-sol10-ifm.tar

One problem with this many inodes is that fsck will use a *LOT* of RAM
to fsck the filesystem and this might start swapping which easily will
cause things to take virtually forever...

You definitiely want to use logging on these filesystems.

(I use this in a fat filesystem that I use as an rsync target from a
number of other servers - so if everything fails I can just newfs
the filesystem and re-rsync everything).

- Peter
--
--
Peter Eriksson <peter@xxxxxxxxxx> Phone: +46 13 28 2786
Computer Systems Manager/BOFH Cell/GSM: +46 705 18 2786
Physics Department, Linköping University Room: Building F, F203
.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Dougii
References:
New 1.4T SVM UFS fs does not have enough inodes
From: Dougii
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Jim Prescott
Prev by Date: Re: ntp in cluster mode
Next by Date: Re: Instal curl
Previous by thread: Re: New 1.4T SVM UFS fs does not have enough inodes
Next by thread: Re: New 1.4T SVM UFS fs does not have enough inodes
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: performance of filesystem xattrs with Samba4
... filesystems at: ... Andreas to add support to ext3 for large inodes.
... and only lost a very small amount when xattrs were not ... clearly how

much
we are paying for journaling. ... (Linux-Kernel)
Re: 20TB Storage System
... > you may not want to create filesystems over 3TB if you want fsck to ...
I don't know if that's using the default newfs settings ... > (which would

create an
insane number of inodes), ... (freebsd-hackers)
Re: 20TB Storage System
... > you may not want to create filesystems over 3TB if you want fsck to ...
I don't know if that's using the default newfs settings ... > (which would

create an
insane number of inodes), ... (freebsd-performance)
Re: 20TB Storage System
... > you may not want to create filesystems over 3TB if you want fsck to ...
I don't know if that's using the default newfs settings ... > (which would

create an
insane number of inodes), ... (freebsd-questions)
Re: 1 Gigabyte SD card "full" after 467 MB [SOLVED]
... filesystems, including ext3, where it's possible to run out of inodes ...

In
practice most filesystems have huge numbers of inodes, ... the real problem

must have been
inodes, directory entries, etc. ... just have nautilus (or konqueror or any

other
application you want to ... (Ubuntu)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(03)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
----
Re: New 1.4T SVM UFS fs does not have enough inodes

-----------------------------------------------------------------------------

---

From: "Dougii" <DouglasDenny@xxxxxxxxx>
Date: 7 Jun 2006 03:58:05 -0700

-----------------------------------------------------------------------------

---
Thanks to Jim and Peter on this. I opened a ticket with Sun and indeed
they told me the limitation for TB filesystems is 1024*1024 / TB. I
will ask sun to remove this limit as part of my ticket, but I will not
hold my breath.

I am going to look at ZFS again, but it is still not clear to me if it
is in general release, i.e. supported under Solaris 10. (3/05 or 1/06).
Maybe it is time to start testing ZFS on some QC systems.

-Doug

Peter Eriksson wrote:

Jim Prescott <jgp@xxxxxxxxxxxxxxxxxx> writes:

In article <1149515332.749153.145950@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Dougii <DouglasDenny@xxxxxxxxx> wrote:

How do I get more inodes in a multi-terabyte file UFS system?

Ask Sun to remove the arbitrary minimum of "nbpi=1048576" for EFI
filesystems. From what I've heard the limit is there because large
filesystems with lots of inodes take a long time to fsck. To me
a potential performance issue might justify a warning message and
a change in defaults, but not a hard limit.

If you can get by with .99T you can get as many inodes as you need.
Based on the df output you presented your current inode density is
around 211k so you could use "newfs -i 100000" on the .99T filesystem
to get a little extra usable space by making fewer inodes.

ZFS doesn't suffer from this limitation. I've also heard that with
the sources from OpenSolaris it is easy to bypass the EFI inode limit.

Yup. It's really simple. You can get a premodified version from:

ftp://ftp.ifm.liu.se/pub/unix/mkfs-sol10-ifm.tar

One problem with this many inodes is that fsck will use a *LOT* of RAM
to fsck the filesystem and this might start swapping which easily will
cause things to take virtually forever...

You definitiely want to use logging on these filesystems.

(I use this in a fat filesystem that I use as an rsync target from a
number of other servers - so if everything fails I can just newfs
the filesystem and re-rsync everything).

- Peter
--
--
Peter Eriksson <peter@xxxxxxxxxx> Phone: +46 13 28 2786
Computer Systems Manager/BOFH Cell/GSM: +46 705 18 2786
Physics Department, Linköping University Room: Building F, F203

.

-----------------------------------------------------------------------------

---

Follow-Ups:
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Darren Dunham
References:
New 1.4T SVM UFS fs does not have enough inodes
From: Dougii
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Jim Prescott
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Peter Eriksson
Prev by Date: Re: Configuration of Solaris OS
Next by Date: Re: Configuration of Solaris OS
Previous by thread: Re: New 1.4T SVM UFS fs does not have enough inodes
Next by thread: Re: New 1.4T SVM UFS fs does not have enough inodes
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: performance of filesystem xattrs with Samba4
... filesystems at: ... Andreas to add support to ext3 for large inodes.
... and only lost a very small amount when xattrs were not ... clearly how

much
we are paying for journaling. ... (Linux-Kernel)
Re: 1 Gigabyte SD card "full" after 467 MB [SOLVED]
... filesystems, including ext3, where it's possible to run out of inodes ...

In
practice most filesystems have huge numbers of inodes, ... the real problem

must have been
inodes, directory entries, etc. ... just have nautilus (or konqueror or any

other
application you want to ... (Ubuntu)
Re: All filesystems change to read-only after a while with kernel 2.6.6-1.435
... > of my filesystems (ext3) change to read-only. ... > about inodes
being members of orphan nodes. ... > that and the subsequent reboot works

fine. ...
(Fedora)
Re: New 1.4T SVM UFS fs does not have enough inodes
... Dougii wrote: ... filesystems with lots of inodes take a long time to
fsck. ... to get a little extra usable space by making fewer inodes. ... One

problem
with this many inodes is that fsck will use a *LOT* of RAM ...

(comp.unix.solaris)
Re: Can I use a round bobbin redirects in a shell ?
... names do not have permissions. ... > A name (struct dirent) is a pointer
to an inode. ... "File serial number", nothing about inodes. ... >> Where
is it stated in POSIX that all filesystems must use inodes? ...

(uk.comp.os.linux)

-----------------------------------------------------------------------------

---

We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(05)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      
-----
Re: New 1.4T SVM UFS fs does not have enough inodes

-----------------------------------------------------------------------------

---

From: Darren Dunham <ddunham@xxxxxxxxxxxxxxxx>
Date: Wed, 07 Jun 2006 20:17:37 GMT

-----------------------------------------------------------------------------

---
Dougii <DouglasDenny@xxxxxxxxx> wrote:

I am going to look at ZFS again, but it is still not clear to me if it
is in general release, i.e. supported under Solaris 10. (3/05 or 1/06).
Maybe it is time to start testing ZFS on some QC systems.

No. It will be included in the 6/06 release.

--
Darren Dunham ddunham@xxxxxxxx
Senior Technical Consultant TAOS http://www.taos.com/
Got some Dr Pepper? San Francisco, CA bay area
< This line left intentionally blank to confuse you. >
.

-----------------------------------------------------------------------------

---

References:
New 1.4T SVM UFS fs does not have enough inodes
From: Dougii
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Jim Prescott
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Peter Eriksson
Re: New 1.4T SVM UFS fs does not have enough inodes
From: Dougii
Prev by Date: multiple processes of mysql/httpd running...
Next by Date: Re: Can't see where is the problem
Previous by thread: Re: New 1.4T SVM UFS fs does not have enough inodes
Next by thread: Users cannot login after changing file permissions
Index(es):
Date
Thread
We are proud to have Web Hosting and Rack Housing from 9 Net Avenue

Deutschland.
(02)
Security UNIX Linux Coding Shop Directory Usenet

Mailing-ListsNewsgroupsAboutPrivacySearch  unix.derkeiler.com  > Newsgroups 

> comp.unix.solaris  > 2006-06      

7、Re: ntp in cluster mode

-----------------------------------------------------------------------------

---

From: Logan Shaw <lshaw-usenet@xxxxxxxxxxxxx>
Date: Tue, 06 Jun 2006 04:59:02 GMT

-----------------------------------------------------------------------------

---
Przem wrote:
I need advice when configuring ntp on cluster.

For now it uses /etc/inet/ntp.conf.cluster and server 127.127.1.0 as time

server.
But because of that this cluster at this moment 12 minutes faster than my

whole environment.

I haven't set up a cluster, although I have used ntp a fair bit.

My advice would be that you need to get your clock calibrated
on every server before you set them up to try to run independently.
The most straightforward way I can think of to do that is to run
ntpd against an external server for a week or so, then let it build
an ntp.drift file that it can use to compensate for the natural
drift of the system's clock.

How can I configure it to use external ntp server?
As I can see, /etc/init.d/xntpd.cluster stopes when it find

/etc/inet/ntp.conf, so I assume I should still use

/etc/inet/ntp.conf.cluster.

Can I simply change field "server 127.127.1.0" to i.e. "server

ntp.server.my"? Will nodes be still synchronized than?

If you have Internet connectivity available, the easiest thing would
be to just synchronize your servers to external servers. If you
really want to ensure that your local machines stay in sync with
each other, you can solve that in one of at least two ways:

(1) synchronize each internal server with enough external servers
that the chances of one of them being far off is really small,
just because it would require too many failures (of external
servers) for that to be likely to happen, or
(2) synchronize one server with external servers, then synchronize
your other servers to it; then your internal servers will not
all be on equal footing because one will be the master, but
they should still stay in sync for a while if you lose internet
connectivity for a day or two.

In general, keep in mind that just setting up a cluster of servers
trying to sync to each other will not give you reliable and
accurate timekeeping. Computers make OK clocks but not good ones,
and in order to keep in sync, you need to ultimately be drawing
your information from a good clock (either from a public ntp server
or from a hardware clock, like a GPS receiver).

My actulaf conf files are (both same):
server 127.127.1.0
peer clusternode1-priv prefer
peer clusternode2-priv
driftfile /var/ntp/ntp.drift
filegen peerstats file peerstats type day enable
filegen loopstats file loopstats type day enable
filegen clockstats file clockstats type day enable

how should they look after I switch to external ntp source?

Just look at what they suggest doing at http://www.pool.ntp.org/ .
If you are in Europe, you should be able to put

server 0.europe.pool.ntp.org
server 1.europe.pool.ntp.org
server 2.europe.pool.ntp.org

and remove all the "peer" lines, and that should probably do it
for you.

- Logan
.

-----------------------------------------------------------------------------

---

References:
ntp in cluster mode
From: Przem
Prev by Date: Re: how to find the total number of open file descriptors per

application
Next by Date: Re: New 1.4T SVM UFS fs does not have enough inodes
Previous by thread: ntp in cluster mode
Next by thread: New 1.4T SVM UFS fs does not have enough inodes
Index(es):
Date
Thread

-----------------------------------------------------------------------------

---

Relevant Pages
Re: Couple of clustering questions
... vmware gsx 2 and build your envisioned cluster environment on that to

test. ...
NT 4.0 servers and are looking to upgrade to 2003 server and want to explore

... that I
want to fail over, but about software that does not need redundancy. ...

nodes at the same
time that I need to divide the SAN into multiple disks. ...

(microsoft.public.windows.server.clustering)
Re: Exch 2003 SP2 - applied on one node, but cant move resources
... resources to Node2, the failover did not complete because 'system

attendant' ...
Virtual Exchange server and failover occurred normally again upon taking ...

cluster
resources oline. ... (microsoft.public.exchange.admin)
Re: Couple of clustering questions
... I have had several issues related to VMWare GSX ... >vmware gsx 2 and

build
your envisioned cluster ... server but does not ... Do all clustered services
that can fail over use ... (microsoft.public.windows.server.clustering)
Re: Basic clustering Q
... Whats wrong with a bit of server consolidation? ... You can never have

enough
DC's to support that cluster! ... >> Hot spare hard disk to provide extra
FT across all disk configs. ... it's about support and having a working

configuration ...
(microsoft.public.windows.server.clustering)
Re: Best Configuration for a 3 Node SQL 2000 Cluster on Windows 2003?
... > I'm not sure if I'm setting up SQL right... ... > cluster, but
I want to make sure we don't shoot ourselves in the foot. ... dedicated SQL

server should
have minimal need for a paging ... During install time you can select which

cluster nodes
to install ... (microsoft.public.sqlserver.clustering)

--------------------------------------------------------------