This site is now 100% read-only, and retired.
#7
Posted by tong on Wed 1 May 2013 at 14:16
It's well known that fail to plan means plan to fail. But when comes to Disk Encryption, I did not see any reasonably planning on disk failure, even though I've googled extensively.
My understanding/impression is that with Full Disk Encryption, even a single bad sector will have a much larger impact than itself and might ruin the whole disk. That's a rather big risk right there, but I haven't found article on how to cope with the problem.
My understanding/impression is that with Full Disk Encryption, even a single bad sector will have a much larger impact than itself and might ruin the whole disk. That's a rather big risk right there, but I haven't found article on how to cope with the problem.
This entry has been truncated read the full entry.
[0 Comments ]
#6
Posted by tong on Tue 29 Aug 2006 at 20:56
[reposted from mlist]
Subject: About the batch command
Newsgroups: gmane.linux.debian.user
Date: Wed, 16 Aug 2006 12:19:50 +0200
Hi
The traditional Unix batch command is old and not sexy enough to generate
enough talk among the Linux users. However, I recently read a post that
hinted the powerfulness of the old utility. To make the long story short,
here are my questions, which are not explained clearly in 'man batch':
- I know that if I list my tasks one by one and sent to the batch command
at once, they'd be executed one by one. But what if I issue 10 different
batch commands one by one, which are all scheduled "now", would they still
be executed one by one? why?
- Which parameter is controlling the sequential execution? What if I want
to execute batch commands two at a time?
- Would batch jobs compete with CPU resources with my foreground tasks?
- Do I need to "nice" the tasks before submitting to batch? If not, what
is the default nice value for batch jobs?
thanks
Subject: About the batch command
Newsgroups: gmane.linux.debian.user
Date: Wed, 16 Aug 2006 12:19:50 +0200
Hi
The traditional Unix batch command is old and not sexy enough to generate
enough talk among the Linux users. However, I recently read a post that
hinted the powerfulness of the old utility. To make the long story short,
here are my questions, which are not explained clearly in 'man batch':
- I know that if I list my tasks one by one and sent to the batch command
at once, they'd be executed one by one. But what if I issue 10 different
batch commands one by one, which are all scheduled "now", would they still
be executed one by one? why?
- Which parameter is controlling the sequential execution? What if I want
to execute batch commands two at a time?
- Would batch jobs compete with CPU resources with my foreground tasks?
- Do I need to "nice" the tasks before submitting to batch? If not, what
is the default nice value for batch jobs?
thanks
[1 Comment ]
#5
Posted by tong on Sat 5 Aug 2006 at 15:17
[reposted from
Newsgroups: gmane.linux.debian.user
Date: Tue, 18 Jul 2006 11:21:17 +0200
]
hi
whenever I add a user in grml (Debian LiveCD, http://grml.org/), I get the following error:
rpcinfo: can't contact portmapper: RPC: Remote system error - Connection
refused
The respond that I got is that
- the Debian's "useradd" tried to run userauth via NIS by default, which
will make rpc calls/connections.
- The reason that I didn't notice this before is because Debian runs
portmapper by default - which grml doesn't.
Although I get the reason, I wasn't able to get the answer to my question
-- is there any way to disable Debian's useradd from using NIS and making
rpc calls/connections?
Please help
thanks
Newsgroups: gmane.linux.debian.user
Date: Tue, 18 Jul 2006 11:21:17 +0200
]
hi
whenever I add a user in grml (Debian LiveCD, http://grml.org/), I get the following error:
rpcinfo: can't contact portmapper: RPC: Remote system error - Connection
refused
The respond that I got is that
- the Debian's "useradd" tried to run userauth via NIS by default, which
will make rpc calls/connections.
- The reason that I didn't notice this before is because Debian runs
portmapper by default - which grml doesn't.
Although I get the reason, I wasn't able to get the answer to my question
-- is there any way to disable Debian's useradd from using NIS and making
rpc calls/connections?
Please help
thanks
[5 Comments ]
#4
Posted by tong on Mon 17 Jul 2006 at 16:04
[reposted from debian-user]
Subject: defoma how to
Newsgroups: gmane.linux.debian.user
Date: Sun, 16 Jul 2006 17:20:02 +0200
Hi
Can you recommend a good defoma how to guide please?
Specifically,
- how to add a truetype font path for defoma to manage?
- If all X-fonts are managed by defoma, theoretically, is it true that I
can leave only the following two FontPath in the "Files" section, while
removing all the rest of FontPath(s)?
FontPath "/var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType" FontPath
"/var/lib/defoma/x-ttcidfont-conf.d/dirs/CID"
Thanks
Subject: defoma how to
Newsgroups: gmane.linux.debian.user
Date: Sun, 16 Jul 2006 17:20:02 +0200
Hi
Can you recommend a good defoma how to guide please?
Specifically,
- how to add a truetype font path for defoma to manage?
- If all X-fonts are managed by defoma, theoretically, is it true that I
can leave only the following two FontPath in the "Files" section, while
removing all the rest of FontPath(s)?
FontPath "/var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType" FontPath
"/var/lib/defoma/x-ttcidfont-conf.d/dirs/CID"
Thanks
[0 Comments ]
#3
Posted by tong on Mon 17 Jul 2006 at 03:09
A collection of mail forwarding options, ranging from simplest satellite systems to normal mail forwarding via exim, qmail, and sendmail, and to the advanced topics like using Postfix to relay email to SMTP server that enforces TLS (eg. nowadays ISP and/or gmail).
If you control satellite systems and you're not expecting to send much mail, and you don't need much processing then running a full mailserver is probably overkill. Perhaps the simplest software to use is the nullmailer program
http://www.debian-administration.org/articles/397
Setting up mail forwarding for exim, qmail, and sendmail
http://www.debian-administration.org/articles/63
how to make postfix to use SMTP auth for sending mail to a smarthost or a
transport map relay server.
http://www.k2.on.lk/fourm/viewtopic.php?t=30
Gmail on Home Linux Box using Postfix
http://souptonuts.sourceforge.net/postfix_tutorial.html
Having read all above, I am still not able to setup my Postfix to relay my mails through my ISP (with TLS required). I had tried many times but all failed. I've also been trying to raise the question at every possible place. But so far respond answers the question.
I really appreciate if anyone can give me a hint. Thanks a lot.
The reason that I need the mail relay is because my ISP has blocked my 25 mail port. So it is impossible for me to file bug reports, use my mail to backup my save-config, etc ... Now that nearly 90% of people are suffering from this mail port block, this feature would be requested again and again.
If you control satellite systems and you're not expecting to send much mail, and you don't need much processing then running a full mailserver is probably overkill. Perhaps the simplest software to use is the nullmailer program
http://www.debian-administration.org/articles/397
Setting up mail forwarding for exim, qmail, and sendmail
http://www.debian-administration.org/articles/63
how to make postfix to use SMTP auth for sending mail to a smarthost or a
transport map relay server.
http://www.k2.on.lk/fourm/viewtopic.php?t=30
Gmail on Home Linux Box using Postfix
http://souptonuts.sourceforge.net/postfix_tutorial.html
Having read all above, I am still not able to setup my Postfix to relay my mails through my ISP (with TLS required). I had tried many times but all failed. I've also been trying to raise the question at every possible place. But so far respond answers the question.
I really appreciate if anyone can give me a hint. Thanks a lot.
The reason that I need the mail relay is because my ISP has blocked my 25 mail port. So it is impossible for me to file bug reports, use my mail to backup my save-config, etc ... Now that nearly 90% of people are suffering from this mail port block, this feature would be requested again and again.
[1 Comment ]
#2
Posted by tong on Wed 28 Jun 2006 at 01:09
[The following message is reposted from gmane.linux.debian.user, which has been sittng there for a while...]
On Sun, 25 Jun 2006 16:19:48 -0400, Tong wrote:
> I got the "pre-installation script returned error" during package
> installation. Is there any solution / workaround? What else should I
> check?
>
> thanks
>
> Tong
>
> PS. here is the full info:
>
> Installing linux-image-2.6.15-1-k7 under etch failed:
>
> % apt-get install linux-image-2.6.15-1-k7 Suggested packages:
> linux-doc-2.6.15 linux-source-2.6.15 lilo grub fdutils
> Recommended packages:
> libc6-i686
> The following NEW packages will be installed:
> linux-image-2.6.15-1-k7
> 0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
> Preconfiguring packages ...
> (Reading database ... 52510 files and directories currently installed.)
> Unpacking linux-image-2.6.15-1-k7 (from
> .../linux-image-2.6.15-1-k7_2.6.15-8_i386.deb) ... Ok, Aborting dpkg:
> error processing
> /var/cache/apt/archives/linux-image-2.6.15-1-k7_2.6.15-8_i386.deb
> (--unpack):
> subprocess pre-installation script returned error exit status 1
> Errors were encountered while processing:
> /var/cache/apt/archives/linux-image-2.6.15-1-k7_2.6.15-8_i386.deb
> E: Sub-process /usr/bin/dpkg returned an error code (1)
I just format the disk and did a clean base system installation then tried
again. Same error.
Installing linux-image-2.6.15-1-686 is broken too:
The following extra packages will be installed:
linux-image-2.6.15-1-686
Suggested packages:
linux-doc-2.6.15 linux-source-2.6.15 lilo grub fdutils
Recommended packages:
libc6-i686
The following NEW packages will be installed:
linux-image-2.6.15-1-686
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded. 2 not fully
installed or removed.
Need to get 0B/15.8MB of archives.
After unpacking 45.1MB of additional disk space will be used. Do you want
to continue [Y/n]?
Preconfiguring packages ...
(Reading database ... 7356 files and directories currently installed.)
Unpacking linux-image-2.6.15-1-686 (from
.../linux-image-2.6.15-1-686_2.6.15-8_i386.deb) ... Ok, Aborting
dpkg: error processing
/var/cache/apt/archives/linux-image-2.6.15-1-686_2.6.15-8_i386.deb
(--unpack):
subprocess pre-installation script returned error exit status 1
Errors were encountered while processing:
/var/cache/apt/archives/linux-image-2.6.15-1-686_2.6.15-8_i386.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)
On Sun, 25 Jun 2006 16:19:48 -0400, Tong wrote:
> I got the "pre-installation script returned error" during package
> installation. Is there any solution / workaround? What else should I
> check?
>
> thanks
>
> Tong
>
> PS. here is the full info:
>
> Installing linux-image-2.6.15-1-k7 under etch failed:
>
> % apt-get install linux-image-2.6.15-1-k7 Suggested packages:
> linux-doc-2.6.15 linux-source-2.6.15 lilo grub fdutils
> Recommended packages:
> libc6-i686
> The following NEW packages will be installed:
> linux-image-2.6.15-1-k7
> 0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
> Preconfiguring packages ...
> (Reading database ... 52510 files and directories currently installed.)
> Unpacking linux-image-2.6.15-1-k7 (from
> .../linux-image-2.6.15-1-k7_2.6.15-8_i386.deb) ... Ok, Aborting dpkg:
> error processing
> /var/cache/apt/archives/linux-image-2.6.15-1-k7_2.6.15-8_i386.deb
> (--unpack):
> subprocess pre-installation script returned error exit status 1
> Errors were encountered while processing:
> /var/cache/apt/archives/linux-image-2.6.15-1-k7_2.6.15-8_i386.deb
> E: Sub-process /usr/bin/dpkg returned an error code (1)
I just format the disk and did a clean base system installation then tried
again. Same error.
Installing linux-image-2.6.15-1-686 is broken too:
The following extra packages will be installed:
linux-image-2.6.15-1-686
Suggested packages:
linux-doc-2.6.15 linux-source-2.6.15 lilo grub fdutils
Recommended packages:
libc6-i686
The following NEW packages will be installed:
linux-image-2.6.15-1-686
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded. 2 not fully
installed or removed.
Need to get 0B/15.8MB of archives.
After unpacking 45.1MB of additional disk space will be used. Do you want
to continue [Y/n]?
Preconfiguring packages ...
(Reading database ... 7356 files and directories currently installed.)
Unpacking linux-image-2.6.15-1-686 (from
.../linux-image-2.6.15-1-686_2.6.15-8_i386.deb) ... Ok, Aborting
dpkg: error processing
/var/cache/apt/archives/linux-image-2.6.15-1-686_2.6.15-8_i386.deb
(--unpack):
subprocess pre-installation script returned error exit status 1
Errors were encountered while processing:
/var/cache/apt/archives/linux-image-2.6.15-1-686_2.6.15-8_i386.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)
[0 Comments ]
#1
Posted by tong on Sat 21 Jan 2006 at 23:08
Will "nice"d background tasks interfere (slow down) normal foreground
task? I used to amaze my Windoze-using friends by burning CD/DVDs while
carrying out my normal work. I normally point to the "top" output,
which says something like "Tasks: 133 total", and say to my astonished
friend, "I can burn my CD/DVDs while there are over a hundred task running
at the same time" -- just enjoy looking at their
astonished faces. However, there is one thing that really make me
question how Linux schedule its tasks.
task? I used to amaze my Windoze-using friends by burning CD/DVDs while
carrying out my normal work. I normally point to the "top" output,
which says something like "Tasks: 133 total", and say to my astonished
friend, "I can burn my CD/DVDs while there are over a hundred task running
at the same time" -- just enjoy looking at their
astonished faces. However, there is one thing that really make me
question how Linux schedule its tasks.
There is this special case, which will let me make coasters every time
-- when nget is updating it headers. I'm using a not-so-monster news
server, which can't compete with supper-monster news server like
giganews, etc. However, whenever nget updates it headers, my computer is
brought down to its knees. I run my nget with "nice -n 12", and my growisofs
(for DVD burning) with "nice -n -2", even so I make coasters from time
to time.
I think it must related with kswapd0 -- the kernel page fault swap
daemon. Shouldn't page faults caused by "nice -n 12" tasks have much lower
priorities then that from "nice -n -2"? I think it should, but why I
make coasters from time to time then?
My box's memory is 768M. I don't think it is too small.
tong
PS.
[4 Comments ]