Nfs Mounting
2007-12-24 21:50:00
Today when i tried nfs mounting, the mount point owner and group
automatically changes to one of our ex-system administrators username
is there any soultion, the mounting was done with root previlages.
Rgds
Sree
----- Original Message -----
From: <codeprof-request at codeprof.com>
To: <codeprof at codeprof.com>
Sent: Wednesday, February 06, 2002 9:01 PM
Subject: codeprof digest, Vol 1 #1035 - 10 msgs
> Send codeprof mailing list submissions to
> codeprof at codeprof.com
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://www.codeprof.com/execute/ask/?codeinfoid=10594
> or, via email, send a message with subject or body 'help' to
> codeprof-request at codeprof.com
>
> You can reach the person managing the list at
> codeprof-admin at codeprof.com
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of codeprof digest..."
>
>
> Today's Topics:
>
> 1. SUMMARY: user password error (Trinh, Linh [C] )
> 2. SUMMARY: Hypothetical Question: LTO Drives into L1000 robot ? (Tim
Chipman)
> 3. customer performance issue... (Paul Wiggins)
> 4. Compaq FCTC-II Support for StorEdge L700 (Aka StorageTek L700)
(Caparroso, Nelson T. (AAS))
> 5. Password Problem on Sparc (Wallace Kha)
> 6. SUMMARY: backing out solaris upgrade (Nate Campi)
> 7. Running out of space in /tmp, weird. (Mark Lin)
> 8. printers epson (Amiri Amar)
> 9. Summary: Running out of space in /tmp - (with one puzzle
> question at the end of this summary) (Mark Lin)
>
> --__--__--
>
> Message: 1
> From: "Trinh, Linh [C] " <TrinhL at nima.mil>
> To: "Trinh, Linh [C] " <TrinhL at nima.mil>,
> "'codeprof at codeprof.com'" <codeprof at codeprof.com>
> Subject: SUMMARY: user password error
> Date: Wed, 6 Feb 2002 10:21:32 -0500
>
> Hello,
>
> It was the path that gave the error. /etc was infront of
/usr/sbin,/usr/bin
> I move the /etc behind them and it works
> The permission was correct
> I received over 20 emails with different troubleshooting
>
> Thank you all,
>
> Linh Trinh
>
> -----Original Message-----
> From: Trinh, Linh [C]
> Sent: Tuesday, February 05, 2002 6:02 PM
> To: codeprof at codeprof.com
> Subject: user password error
>
>
> Hi,
>
> I have a user that can not changing his own password.
> When he login with his user account and executing the command
> %passwd
> /etc/passwd: cannot open
>
> I can change mine as regular user.
> Please give some suggestions
>
> TIA,
>
> Linh Trinh
>
> --__--__--
>
> Message: 2
> Date: Wed, 06 Feb 2002 11:23:50 -0500
> From: Tim Chipman <chipman at ecopiabio.com>
> To: codeprof at codeprof.com
> Subject: SUMMARY: Hypothetical Question: LTO Drives into L1000 robot ?
>
> Thanks to everyone who replied (Troy Abernathy, "thetrick", Anil
> Sreedharan, Charles Rawls, Edmond van As, Jeff Kennedy, Steve
> Mickeler). There was a wide range of thoughts submitted on the matter.
>
> Bottom line: I'm doing nothing for now, trying to learn a bit more. (I'm
> not holding my breath though). Much feedback kindly drew my attention to
> the fact that DLT, LTO are not absolutely identical in physical
> dimensions. A number of replies indicated that it was quite possibly a
> "doable" thing, depending on a number of unknowns (primarily, the
> ability to tell the robot the new physical spec for the tapes it is
> grabbing ; possibly requiring an upgrade of the "gripper unit" itself.).
>
> Greater detail is given below for those who are into reading about this
> sort of thing.
>
>
> --Tim Chipman
>
>
> -> DLT and LTO are not the same width, height or capacity
>
> [indeed, quite my goal - if they were the same capacity, then there is
> little point in this sort of upgrade :-) ] -- but the real issue here,
> the dimensions, compare thus:
>
> LTO: 4.02" x 4.15" x 0.85"
> DLT: 4.16" x 4.15" x 1.00"
>
> -> apparently it should be "doable" but not trivial ; may require
> upgrade of the "gripper" unit that physically grabs the tapes ; likely
> requires some means of telling the robot firmware the new dimensions of
> the tapes. I wasn't intending to keep DLT support in the upgrade (ie,
> mixed media library) but apparently this reply suggests that it might be
> feasible:
>
> "Not insane nor impossible, but not recommended. I have an ATL P6000
> and I know for a fact it can be done, mixed media capability.However,one
> reason it's not recommended is that you need to slice and dice the
> library. Media slots must be logically partitioned to belong to 2
> different media sets. Doable but not easy to spec out as you may not
> know how many LTO media you need to handle the data you plan to send to
> it. Another is mixing the media in wrong slots. If you or someone else
> accidentally put LTO media in a DLT slot very bad things will happen, to
> the point where you may have to call onsite support to fix your
> robot..... And finally is whether or not you need to upgrade your robot
> to handle both types of media. Even with SDLT I need to have a
> 'universal gripper' to handle both."
>
>
> -> In keeping with the "affirmative" voices: two replies suggested
> (basically) that the robot library itself knows nothing about tapes
> other than their physical property (size, location in the bank). So long
> as different dimensions (height,width) for tapes could be accounted for
> (?somehow?) then the robot really should not care.
>
> "The changer / library has no consideration of tape size. When it hits
> the end of tape, or is issued a dismount, it pulls the tape out
> and replaces it with the next tape.
>
> As long as the tape drive itself is recognized by Solaris, you should
> have no issues. I've done this type of thing in the 8mm world,
> the big PITA is when the mounts don't quite line up right ..."
>
> and ...
>
> "They may be the same in x-y directions, but they are definately only
> 2/3 as thick. The picker would have to be able to understand the
> difference. It is possible to reprogram some libraries (older
> E-mass libraries, for instance), but as to yours specifically,
> I don't know."
>
>
> -> One person suggested to buy a reburb L3500 (for example) with DLT as
> a means of getting capacity scaling on backups, rather than investing
> more in the L1000 robot. (ie, upgrade 4 SDLT drives into an L1000 might
> cost upwards of $40k usd, vs. a refurb L3500 with 7 x DLT7000 drives for
> ~$20k USD).
>
> I'm not at all certain we prefer to invest in old gear or not. Currently
> the price on L20-type units (or the HP gear they are derived from) - a
> dual-LTO-drive, 20-tape robot - has quite an attractive look to it. The
> capacity and throughput are very decent.
>
>
>
> ORIGINAL POSTING:
>
> This is likely a flight of folly, but possibly worth asking. No better
> resoure than this list I can think of for such a query :-)
>
> I'm curious if it is even concievable to put LTO drives into an L1000
> robot (which currently has DLT drives in it). [AFAIK the LTO and DLT
> tapes are physically ~identical hence the robot itself shouldn't be too
> bothered by this aspect of the change?.. however, it probably would
> require the tape drive physical specs to be identical as well though,
> which might be a bit more tricky I suspect..?]
>
> I'm quite certain that quantum support putting *superDLT* into the
> L1000 (although sun doesn't support this yet) .. but I'm curious
> about LTO instad of sDLT for this kind of crazy upgrade.
>
> Given the increased throughput and capacity of LTO drives over DLT, it
> would be a rather nice upgrade path if it were to be an option.
>
> If anyone has any ideas on the feasability of this, I'd certainly
> appreciate hearing your thoughts. [ie, "insane" ; "I've done it" ;
> etc..]
>
>
> Tim Chipman
>
> --__--__--
>
> Message: 3
> From: "Paul Wiggins" <pwiggins at cisco.com>
> To: <codeprof at codeprof.com>
> Cc: "Paul Wiggins" <pwiggins at cisco.com>
> Subject: customer performance issue...
> Date: Wed, 6 Feb 2002 11:24:51 -0500
>
> I'm jumping into critical customer issue where their current Webserver is
> maxing out. The obvious cure is to have them upgrade from the U-10/256 to
> something larger...
>
> Where I'm having a difficult time is I'm new to SUN performance and
> scalability characterization and looking for the top few tools folks 1st
> grab to analyze the current impact of the server load.
>
> More to the point... perfmeter and vmstat show a box that is simply
running
> out of CPU cycles. Where I get confused is the CPU will max out (as
reported
> by perfmeter) regardless of the level of load. Starting the Java based
> webserver causes the CPU to max out... Whether there is 100% or 50% of
> successful load (100% requests eventually serviced)... the perfmeter
reports
> CPU max'd out.
>
> So... it's easy enough to determine disk space and ram needed per 1000
> users... but how do you assess the benefit of faster CPU's; when you only
> data that is time based (on a U-10; 1000 users are processed in 30sec,
10000
> users are processed in 333sec; how do you map that to a faster processor)?
>
> Apologies if this is off topic and I realize this is a big question...
just
> looking for a compass heading.
>
> thanks.
> Paul Wiggins
>
> --__--__--
>
> Message: 4
> From: "Caparroso, Nelson T. (AAS)" <Nelson.T.Caparroso at sbcdo.com>
> To: "'codeprof at codeprof.com'" <codeprof at codeprof.com>
> Subject: Compaq FCTC-II Support for StorEdge L700 (Aka StorageTek L700)
> Date: Wed, 6 Feb 2002 10:39:40 -0600
>
> I've got a Compaq FCTC-II (oem'd Crossroads 4xxx?) - no longer in
> production, hence no more firmware upgrades. Does anyone have any
experience
> with this SCSI-FC bridge (Comapq calls ut FC Tape Controller) and more
> specifically with DLT7000 drives on an L700 Library?
>
> --__--__--
>
> Message: 5
> Date: Wed, 6 Feb 2002 08:48:19 -0800 (PST)
> From: Wallace Kha <Wallace.Kha at Sun.COM>
> Reply-To: Wallace Kha <Wallace.Kha at Sun.COM>
> Subject: Password Problem on Sparc
> To: codeprof at codeprof.com
>
> Hi People
> I made a bit of a mistake when I edited the /etc/passwd file by changing
the
> root account to "iroot" (damn vi). So now I can't even do an "su" to
remodify
> the /etc/passwd file. I've tried rebooting and thats even worst, coz I
can't
> even get to the login screen. It gets to the part asking me to hit Ctrl-D
for
> system maintenance or enter for normal startup, but I can't seem to get
past
> that.
>
> Can anyone help me out? this is just a simple workstation (sparc 5 I think
it
> is)
> Much Appreciated
>
> --__--__--
>
> Message: 6
> Date: Wed, 6 Feb 2002 08:53:23 -0800
> From: Nate Campi <nate at campin.net>
> To: codeprof <codeprof at codeprof.com>
> Subject: SUMMARY: backing out solaris upgrade
>
> The first few responses was to look at LiveUpgrade from Sun, which looks
> good. One person asked if I'd really trust a backout procedure anyways,
> which was a good point, and to use conventional backups. A couple
> people asked if it was mirrored, and if so to break the mirror before
> upgrade. A couple more agreed that conventional backups (local tape,
> ufsdump) is best.
>
> I like the mirror idea - the disk isn't mirrored but it's a UE2 and the
> second disk slot isn't used. I'll mirror it with DiskSuite and break the
> mirror before the upgrade.
>
> Now: how do I get the 2.6 DiskSuite software when I don't have the 2.6
> CD's? All I have around are Solaris 7 and 8 CDs. Any pointers welcome.
>
> Thanks to:
>
> Evans, Tim
> TRUCKS, JESSE
> Cho, John
> Frederiksen, Paul
> David Foster
> David Billett
> system administration account <sysadmin at astro.su.se>
> Granzow, Doug
> Cawley,Stephen
> --
> Nate
>
> Without C, We would only have Pasal, Basi, and obol
>
> --__--__--
>
> Message: 7
> From: "Mark Lin" <sv650 at macysadv.com>
> To: <codeprof at codeprof.com>
> Subject: Running out of space in /tmp, weird.
> Date: Wed, 6 Feb 2002 12:11:07 -0500
>
> Hi, I have a urgent problem. My tmp space is running out. But it looks
> like the server forgot to mount the correct swap slice. How can I change
> this without reboot?
>
> thank you
> Mark
>
> here's the info:
> df -k:
> Filesystem kbytes used avail capacity Mounted on
> /dev/dsk/c0t0d0s0 490097 79588 361500 19% /
> /dev/dsk/c0t0d0s5 3579989 858488 2685702 25% /usr
> /proc 0 0 0 0% /proc
> fd 0 0 0 0% /dev/fd
> mnttab 0 0 0 0% /etc/mnttab
> /dev/dsk/c0t0d0s4 482824 295003 139539 68% /var
> swap 16392 8 16384 1% /var/run
> swap 16784 400 16384 3% /tmp
> /dev/dsk/c0t0d0s6 3579989 2462175 1082015 70% /opt
>
>
> /etc/vfstab:
> #device device mount FS fsck mount
> mount
> #to mount to fsck point type pass at boot
> options
> #
> #/dev/dsk/c1d0s2 /dev/rdsk/c1d0s2 /usr ufs 1 yes -
> fd - /dev/fd fd - no -
> /proc - /proc proc - no -
> /dev/dsk/c0t0d0s1 - - swap - no -
> /dev/dsk/c0t0d0s0 /dev/rdsk/c0t0d0s0 / ufs 1 no
> -
> /dev/dsk/c0t0d0s5 /dev/rdsk/c0t0d0s5 /usr ufs 1 no
> -
> /dev/dsk/c0t0d0s4 /dev/rdsk/c0t0d0s4 /var ufs 1 no
> -
> /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /opt ufs 2
yes
> -
> swap - /tmp tmpfs - yes -
>
> --__--__--
>
> Message: 8
> From: Amiri Amar <Amar.Amiri at cnes.fr>
> To: "'codeprof at codeprof.com'" <codeprof at codeprof.com>
> Subject: printers epson
> Date: Wed, 6 Feb 2002 17:09:08 +0100
>
> Hi gurus,
>
> I want to install an epson Aculaser C8500 printer under solaris 7. For the
epson there is no driver for unix.
> I don't know how to configure this printer in order to use multiple tray
(one for the A4 paper and the other one for the A3 paper).
> When I print since a client, by using the option A3, It does'nt work. But
it's ok since the server.
> Could someone help me.
>
> Thanks in advance.
>
> --__--__--
>
> Message: 9
> From: "Mark Lin" <sv650 at macysadv.com>
> To: <codeprof at codeprof.com>
> Subject: Summary: Running out of space in /tmp - (with one puzzle
> question at the end of this summary)
> Date: Wed, 6 Feb 2002 13:04:07 -0500
>
> Thanks for the super fast replies. 16 response in 20 minutes.
>
> Here's the summary:
> Reason for the problem: There is some run away process that eats up all
the
> memory. I didn't find out which process before I reboot, I'll keep an eye
> on it in the future.
>
> Two ways to solve the problme without reboot:
> 1. use the swap command.
> swap -l ->to show my swap slice.
> swap -a /dev/dsk/c0t0d0s1 -> to add my swap slice
>
> 2. use mkfile then swap like this:
> mkfile 500m /opt/swapfile
> swap -a /opt/swapfile
> (dont forget to delete it after reboot)
>
> Afterthoughts:
> my df -lk still looks weird.
> here's my `swap -l` output:
> swapfile dev swaplo blocks free
> /dev/dsk/c0t0d0s1 32,1 16 1052144 1052144
>
> but my df -k looks like this:
>
> /dev/dsk/c0t0d0s0 490097 79444 361644 19% /
> /dev/dsk/c0t0d0s5 3579989 843419 2700771 24% /usr
> /proc 0 0 0 0% /proc
> fd 0 0 0 0% /dev/fd
> mnttab 0 0 0 0% /etc/mnttab
> /dev/dsk/c0t0d0s4 482824 263233 171309 61% /var
> swap 1010792 8 1010784 1% /var/run
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> swap 1010792 8 1010784 1% /tmp
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> /dev/dsk/c0t0d0s6 3579989 2493988 1050202 71% /opt
>
> I only have 513MB assigned for swap, and 640MB of RAM, why would I have
two
> swap spaces mounted on two place, /var/run and /tmp? Can someone tell me
> when does /var/run get mounted? I dont have that entry in my /etc/vfstab.
>
> thanx for the replies
> Mark
>
> original post:
> --------------------
> Hi, I have a urgent problem. My tmp space is running out. But it looks
> like the server forgot to mount the correct swap slice. How can I change
> this without reboot?
>
> thank you
> Mark
>
> here's the info:
> df -k:
> Filesystem kbytes used avail capacity Mounted on
> /dev/dsk/c0t0d0s0 490097 79588 361500 19% /
> /dev/dsk/c0t0d0s5 3579989 858488 2685702 25% /usr
> /proc 0 0 0 0% /proc
> fd 0 0 0 0% /dev/fd
> mnttab 0 0 0 0% /etc/mnttab
> /dev/dsk/c0t0d0s4 482824 295003 139539 68% /var
> swap 16392 8 16384 1% /var/run
> swap 16784 400 16384 3% /tmp
> /dev/dsk/c0t0d0s6 3579989 2462175 1082015 70% /opt
>
>
> /etc/vfstab:
> #device device mount FS fsck mount
> mount
> #to mount to fsck point type pass at boot
> options
> #
> #/dev/dsk/c1d0s2 /dev/rdsk/c1d0s2 /usr ufs 1 yes -
> fd - /dev/fd fd - no -
> /proc - /proc proc - no -
> /dev/dsk/c0t0d0s1 - - swap - no -
> /dev/dsk/c0t0d0s0 /dev/rdsk/c0t0d0s0 / ufs 1 no
> -
> /dev/dsk/c0t0d0s5 /dev/rdsk/c0t0d0s5 /usr ufs 1 no
> -
> /dev/dsk/c0t0d0s4 /dev/rdsk/c0t0d0s4 /var ufs 1 no
> -
> /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /opt ufs 2
yes
> -
> swap - /tmp tmpfs - yes -
> _______________________________________________
> codeprof mailing list
> codeprof at codeprof.com
> http://www.codeprof.com/execute/ask/?codeinfoid=10594
>
>
> --__--__--
>
> _______________________________________________
> codeprof mailing list
> codeprof at codeprof.com
> http://www.codeprof.com/execute/ask/?codeinfoid=10594
>
>
> End of codeprof Digest
Comments
Got something to say?
You must be logged in to post a comment.
