Nabble - Debian ISP

kredit tanpa agunan 8-125 jt..adm/provisi 1,5 % saja

2008-10-03T19:46:47Z

Pinjaman uang tunai tanpa agunan dari bank terbesar di asia ,memberikan kemudahan bagi anda untuk

mendapatkannya dengan persyaratan yang sangat mudah dan flexible cukup dengan fc ktp dan

fc credit card (sisi depannya saja),fc billing tagihan 1 bulan terakhir (bila ada)

dan bunga yang sangat kompetitif ,proses hingga pencairan hanya membutuhkan MAXIMAL

5- 7 hari kerja saja.kemudahan pembayaran yg dapat dilakukan via ATM BCA,LIPPO,BII

atau pun dapat juga lewat kantor pos.;;provisi/bea adm hanya 1,5% dr nilai kredit yg di setujui.

DIBANTU HINGGA CAIR........

hubungi :

windi

94624222

or

"wonderwindi09@gmail.com"

PERHITUNGAN BUNGA/BULAN

TENOR

Jml.Pinjaman 12 bulan 24 bulan 36 bulan

8.000.000 - 14.999.999 1.87% 1.87 %. -

15.000.000 - 125.000.000 1.70% 1.75 %. 1.80 %.

TABEL ANGSURAN

PINJAMAN 1 TAHUN 2 TAHUN 3 TAHUN

8,000,000 816,267 482,933

9,000,000 918,300 543,300

10,000,000 1,020,334 603,666

10,500,000 1,071,350 633,850

11,000,000 1,122,367 664,033

12,000,000 1,224,401 724,400

13,000,000 1,326,434 784,766

14,000,000 1,428,467 845,133

15,000,000 1,505,000 887,500 686,667

16,000,000 1,605,333 946,667 732,444

17,000,000 1,705,667 1,005,833 778,222

18,000,000 1,806,000 1,065,000 824,000

19,000,000 1,906,333 1,124,167 869,778

20,000,000 2,006,667 1,183,333 915,556

21,000,000 2,107,000 1,242,500 961,333

22,000,000 2,207,333 1,301,667 1,007,111

23,000,000 2,307,667 1,360,833 1,052,889

24,000,000 2,408,000 1,420,000 1,098,667

25,000,000 2,508,333 1,479,167 1,144,444

25,500,000 2,558,500 1,508,750 1,167,333

26,000,000 2,608,667 1,538,333 1,190,222

27,000,000 2,709,000 1,597,500 1,236,000

28,000,000 2,809,333 1,656,667 1,281,778

29,000,000 2,909,667 1,715,833 1,327,556

30,000,000 3,010,000 1,775,000 1,373,333

31,000,000 3,110,333 1,834,167 1,419,111

32,000,000 3,210,667 1,893,333 1,464,889

33,000,000 3,311,000 1,952,500 1,510,667

34,000,000 3,411,333 2,011,667 1,556,444

35,000,000 3,511,667 2,070,833 1,602,222

36,000,000 3,612,000 2,130,000 1,648,000

37,000,000 3,712,333 2,189,167 1,693,778

38,000,000 3,812,667 2,248,333 1,739,556

39,000,000 3,913,000 2,307,500 1,785,333

40,000,000 4,013,333 2,366,667 1,831,111

41,000,000 4,113,667 2,425,833 1,876,889

42,000,000 4,214,000 2,485,000 1,922,667

43,000,000 4,314,333 2,544,167 1,968,444

44,000,000 4,414,667 2,603,333 2,014,222

45,000,000 4,515,000 2,662,500 2,060,000

46,000,000 4,615,333 2,721,667 2,105,778

47,000,000 4,715,667 2,780,833 2,151,556

48,000,000 4,816,000 2,840,000 2,197,333

49,000,000 4,916,333 2,899,167 2,243,111

50,000,000 5,016,667 2,958,333 2,288,889

55,000,000 5,518,333 3,254,167 2,517,778

60,000,000 6,020,000 3,550,000 2,746,667

65,000,000 6,521,667 3,845,833 2,975,556

70,000,000 7,023,333 4,141,667 3,204,444

75,000,000 7,525,000 4,437,500 3,433,333

80,000,000 8,026,667 4,733,333 3,662,222

85,000,000 8,528,333 5,029,167 3,891,111

90,000,000 9,030,000 5,325,000 4,120,000

95,000,000 9,531,667 5,620,833 4,348,889

100,000,000 10,033,333 5,916,667 4,577,778

105,000,000 10,535,000 6,212,500 4,806,667

110,000,000 11,036,667 6,508,333 5,035,556

115,000,000 11,538,333 6,804,167 5,264,444

120,000,000 12,040,000 7,100,000 5,493,333

125,000,000 12,541,667 7,395,833 5,722,222

help with the workaround.org postfix isp style email setup.

2008-09-30T05:32:12Z

I have setup my email like the configuration recommended on
workaround.org tutorial.
But having a couple of issues and hope you can help.

1. i get a lot of connection refused errors, so i modified by
$mynetworks line is postfix to include the loopback address, public
and private ip addresses.
But i do still get the error of connection refused from time to time.

2. When i send email to yahoo,live/hotmail,and gmail - it all goes to
the spam bucket --do not know why - I have setup a reverse dns record
for my domain name.
Is there anything else i can check ???

3. I get an error :
Sep 30 08:16:05 rider postfix/smtpd[9521]: warning: cannot get
certificate from file /etc/ssl/certs/postfix.pem
Sep 30 08:16:05 rider postfix/smtpd[9521]: warning: TLS library
problem: 9521:error:02001002:system library:fopen:No such file or
directory:bss_file.c:352:fopen('/etc/ssl/certs/postfix.pem','r'):
Sep 30 08:16:05 rider postfix/smtpd[9521]: warning: TLS library
problem: 9521:error:20074002:BIO routines:FILE_CTRL:system
lib:bss_file.c:354:
Sep 30 08:16:05 rider postfix/smtpd[9521]: warning: TLS library
problem: 9521:error:140DC002:SSL
routines:SSL_CTX_use_certificate_chain_file:system lib:ssl_rsa.c:720:
Sep 30 08:16:05 rider postfix/smtpd[9521]: cannot load RSA certificate
and key data
Sep 30 08:16:05 rider postfix/smtpd[9521]: connect from localhost[127.0.0.1]

THis error is killing me becasue i create the file as specified in the
tutorial and the file disappears...is there something i am doing wrong
here ????

thanks...need all the hlp i can get.

thanks
mjh

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T12:54:45Z

PÁSZTOR György wrote:
> Hi,
>
> "Nate Duehr" <nate@...> írta 2008-09-29 01:35-kor:
>> Sure seems a lot easier to built a proper Spanning-Tree hierarchy that
>> blocks only when necessary and unblocks automatically when the
>> condition clears (no "err-disabled" if the problem has cleared itself)?
> I want to know if someone do sg. like this, and I want to talk with.
> If it clears itself, the problem just "disappears", and stronger the beleif
> that the network is unreliable.

That's why you monitor and you go find root-cause.

>> As far as "hacks" on the network. All comments regarding hard-coding
>> a port ASSUME that the network admin actually knows how to keep and
>> use proper documentation techniques, and is also MONITORING such
>> things via SNMP from the switches, to note inappropriate speed/rate
>> changes.
> Thats nice in theory. The only issue is that, I never seen this nice
> practice.

So you're saying your practice is bad? Time to get control of it.

>> In a properly documented network, hard-setting or autonegotiation
>> aren't "hacks" they're just another setting in the documentation and
>> monitoring systems.
> It's true in a case where you have no choice about incorporating a
> device which doesn't complain standards. But as I said: If a networking
> device doesn't complain so obvious standard, like autonegoiation, it will
> mess up other things too.

That's not a logical argument. Code isn't that way.

That's like saying if there's bugs in Apache for one module, there will
certainly be bugs in another. Non-logical assumption.

People try too hard to associate patterns of "behavior" of a piece of
gear to real bugs. Bugs are bugs. You either have them or you don't.

You can't extrapolate a bug rate from a whole product because a
particular device has a single bug.

What you CAN logically say is that ALL devices have some sort of bug
somewhere in their code. That one's pretty solid, unless you've tied
bugs to money somewhere, and for every bug your vendor has to pay you.

That hasn't happened in computer technology yet. (It has in other
professions, in the form of malpractice lawsuits and insurance.
Doctors, Lawyers, Civil Engineers... when we start having LAWSUITS over
bad code, life will be far more interesting and quality levels will go up.)

> I glanced at your other mail, and I saw, that you wrote about cisco-sun
> difficulties. I think thats a good example: If you have no other working
> solution, It can be the right way. But I have to repeat myself: Only if you
> have proper documentation about those hardcodings/hacks/whatever.

Yes, we agree here. Documentation and monitoring are the network
admin's job.

Nate

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T06:48:57Z

2008/9/29 PÁSZTOR György <pasztor@...>:
> Hi!
>
[snip]
> It sounds like the most common issue of the fixed port:
> Device A sets its FE to fixed 100 FD
> Device B sets its FE to autoneg.
> B try to autoneg: A has disabled it, so doesn't answer.
> B sense the speed and sets 100-HD.
> -> B measures many-many collisions because tries Half Duplex operation on a
> link, where the oppenent part operates in Full.

that was the issue, they configured the cat with fixed port and we didn't know!
BUT that was not a problem when, instead of a linux server, there was connected
a c7200 on the same port of the switch ...

> But, if it's the case, what I suppose, it's not the fault of your cisco
> devices, or your computers. It's a simple PEBKAC.

obviously it's not equipment fault, it's just a matter of configuration,
knowing when a port is fixed and when not...
but sometimes you just *can't* know how some 'computers' are configured,
and in our experience when the connection is cisco <-> cisco aren't
any problems, when it's cisco <-> other equipment fixing the port on the
non-cisco server/switch/whatever has always fixed collision problems
while rarely tweaking on cisco config has...

btw, never had a problem with Gbit connections, only with 100 Mbit, (and
ethtool even refuses to turn autoneg off on GbE fortunately ;)

cheers

--
------------------------------------------------
Daniele Santi .o.
daniele@... ..o () ascii ribbon campaign
Linux User #415108 ooo /\ www.asciiribbon.org
------------------------------------------------

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T06:15:34Z

W dniu 29 września 2008 15:13 użytkownik PÁSZTOR György
<pasztor@...> napisał:
> But, if it's the case, what I suppose, it's not the fault of your cisco
> devices, or your computers. It's a simple PEBKAC.

exactly, we hade those pebkac problems with our local telco, but the
problem existed between keyboard and cisco not 'computer' ;)

regards.

--
Wojciech Ziniewicz
Unix SEX :{look;gawk;find;sed;talk;grep;touch;finger;find;fl
ex;unzip;head;tail; mount;workbone;fsck;yes;gasp;fsck;more;yes;yes;eje
ct;umount;makeclean; zip;split;done;exit:xargs!!;)}

Re: load balanced nic

2008-09-29T06:13:00Z

Hi!

"Mr Shunz" <mrshunz@...> írta 2008-09-29 13:26-kor:
> we have a cat 2950 connected to a server running quagga.
> We kept seeing collision both on the switch (led blinking yellow) and
> on the server (ifconfig from console and sh int eth1 from vtysh). Talking
> with our upstrem provider (it's their cat) they told to fix our port
> and *bam* no more collision in two years. After that we had to temporarily
> replace a linksys managed switch with a cheap one and our 7200 began to show
> collisions on the fE interface which disappeared when we put back the
> linksys and fixed the port there ...
It sounds like the most common issue of the fixed port:
Device A sets its FE to fixed 100 FD
Device B sets its FE to autoneg.
B try to autoneg: A has disabled it, so doesn't answer.
B sense the speed and sets 100-HD.
-> B measures many-many collisions because tries Half Duplex operation on a
link, where the oppenent part operates in Full.

But, if it's the case, what I suppose, it's not the fault of your cisco
devices, or your computers. It's a simple PEBKAC.

Cheers,
gyu

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T05:59:56Z

Another case of fixing a port speed:
We had a 10 base T network (but with 5 and 5E cabling). all went fine for
some years. at a certain time the number of workstations connected were so
much increased that the network became unusable.
We bought some Gigabit switches and as pheripheric swicthes, to conenct
all WS in a room, some chap 100 MB switches. all was fine ... for just a
few weeks ... given more bandwidth even users increased their usage, so
even the 100 MB switch proved unusable ... ot at least it looked so. We
bought a further gigabit one.
Actually it fixed the problem for "local" (that is between hosts in
the same romm) file transfer but speed decreased for "outside" connection,
op to levers worse than when used the 10 MB switch.
Since i had doubt on one of the GB switch i swapped that one for a
knownt-to-work 100 MB one. Performance retuyrned good.
I inquired and discovered that on the cable from the backbone to the romm
having problem (a long and hard to reach cable) 2 wires were defective
somewhere . Changing the wire was out of concern, so the solution was to
set the speed at 100 MB on that junction, so it used only two
"good" pairs, while on GB setting it tried to use also the two defective
ones.

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T04:26:00Z

Hi,

2008/9/28 PÁSZTOR György <pasztor@...>:
> Hi,
[snip]

> Could you show us some proof for this, otherwise I have to assume, that
> this is a nonsense urban legend.
> I have tons of 2950s and 3750s at my workplace which is maintained by me,
> but there is no fixed ports in the whole building.

we have a cat 2950 connected to a server running quagga.
We kept seeing collision both on the switch (led blinking yellow) and
on the server (ifconfig from console and sh int eth1 from vtysh). Talking
with our upstrem provider (it's their cat) they told to fix our port
and *bam* no more collision in two years. After that we had to temporarily
replace a linksys managed switch with a cheap one and our 7200 began to show
collisions on the fE interface which disappeared when we put back the
linksys and fixed the port there ...

cheers

--
------------------------------------------------
Daniele Santi .o.
daniele@... ..o () ascii ribbon campaign
Linux User #415108 ooo /\ www.asciiribbon.org
------------------------------------------------

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T02:15:32Z

Hi,

"Nate Duehr" <nate@...> írta 2008-09-29 01:35-kor:
> Sure seems a lot easier to built a proper Spanning-Tree hierarchy that
> blocks only when necessary and unblocks automatically when the
> condition clears (no "err-disabled" if the problem has cleared itself)?
I want to know if someone do sg. like this, and I want to talk with.
If it clears itself, the problem just "disappears", and stronger the beleif
that the network is unreliable.
Otherwise: If you plan well, you will know at the planning state, what
ports will be backbone ports, and you allow pdu's there, and you will know
what ports will be access ports, and you allow spanning-tree portfast on
them, but if you don't want to mess up things, you should also allow
bpduguard also.

> >Yes! That's another correct observation! If a device doesn't comply a
> >standard which is so obvios, than what can do wrong too?
> >And the other observation is also correct: If you do too many hacks
> >on your
> >network, it's usually undocumented, and it cause many other problems
> >when
> >you connect devices which works correctly, and comply the standards.
>
> As far as "hacks" on the network. All comments regarding hard-coding
> a port ASSUME that the network admin actually knows how to keep and
> use proper documentation techniques, and is also MONITORING such
> things via SNMP from the switches, to note inappropriate speed/rate
> changes.

Thats nice in theory. The only issue is that, I never seen this nice
practice.

> In a properly documented network, hard-setting or autonegotiation
> aren't "hacks" they're just another setting in the documentation and
> monitoring systems.
It's true in a case where you have no choice about incorporating a
device which doesn't complain standards. But as I said: If a networking
device doesn't complain so obvious standard, like autonegoiation, it will
mess up other things too.
I glanced at your other mail, and I saw, that you wrote about cisco-sun
difficulties. I think thats a good example: If you have no other working
solution, It can be the right way. But I have to repeat myself: Only if you
have proper documentation about those hardcodings/hacks/whatever.

Cheers,
gyu

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T00:35:11Z

On Sep 28, 2008, at 10:41 AM, PÁSZTOR György wrote:

> <slightly-offtopic>
> Common problem, when some moron come to here with bridging enabled in
> his/her notebook, and connect to an access port, and then that
> alternative
> OS, whichs name is not spoken - the evil one :-D - starts to send PDUs
> towards the access port, and then it's change its state to err-
> disabled.
> When they come to me for complain about the unreliable network, I tell
> them, what they did: From my viewpoint, they accross the network
> policy, so
> they can be happy, that I still allow them to connect, and re-allow
> the
> err-disabled port.
> </slightly-offtopic>

Sure seems a lot easier to built a proper Spanning-Tree hierarchy that
blocks only when necessary and unblocks automatically when the
condition clears (no "err-disabled" if the problem has cleared itself)?

> Yes! That's another correct observation! If a device doesn't comply a
> standard which is so obvios, than what can do wrong too?
> And the other observation is also correct: If you do too many hacks
> on your
> network, it's usually undocumented, and it cause many other problems
> when
> you connect devices which works correctly, and comply the standards.

As far as "hacks" on the network. All comments regarding hard-coding
a port ASSUME that the network admin actually knows how to keep and
use proper documentation techniques, and is also MONITORING such
things via SNMP from the switches, to note inappropriate speed/rate
changes.

If you're not in control of your network BEFORE messing with hard-
coding things, OF COURSE you'll find this to be a "hack" later when
your brain can't remember what you did years ago.

In a properly documented network, hard-setting or autonegotiation
aren't "hacks" they're just another setting in the documentation and
monitoring systems.

--
Nate Duehr
nate@...

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-29T00:26:17Z

On Sep 28, 2008, at 4:20 AM, Shane Chrisp wrote:
> It is quite common practise to set the ports to what you need so that
> the auto-negotiation does not get it wrong, in particular setting half
> duplex instead of full duplex. This is not so common when its
> "cisco connected to cisco", but very common when its "cisco
> connected to
> some other brand".

Cisco / Sun combinations are PARTICULARLY notorious for this
problem. They have never played nicely between their different forms
of auto-negotiation.

Hard-coding in the server room/closet, auto-negotiation for the PC's
on the general population network is VERY common. Hard-coding in a
data-center environment is also quite common.

One place where this falls down are fiber interfaces. The fiber specs
typically do things "right" without messing with them, and messing
with hard-setting fiber interfaces often leads to other trouble (human
error).

Hard-setting in cases where Ethernet distance lengths for higher
speeds have been exceeded can also be useful. The cable in use may
not consistently work well at a higher speed over long distances, but
will handle lower speed Ethernet just fine.

(Example: Very old buildings with Cat 3 type cabling in risers often
exhibits bad CRC error rates at 100 full. 100 half, or even 10 full/
half hard-set in the switch closet may save a small company struggling
for cash enough money and time to "get by" until the building's
distribution frame cabling can be upgraded. It's a stupid thing to
have to do, but it's better than thousands upon thousands of CRC
errors from old wiring counting up in all the switches.)

In short -- if anyone hasn't been around Ethernet long enough to have
seen times when hard-set settings and auto-negotiation can BOTH cause
trouble, they haven't seen it all yet. I have.

Try a call-center wired with Cat 3 25-pair in riser bundles to 110
punch blocks (old AT&T building) to Cat 5 STP (shielded, not
unshielded twisted pair) installed incorrectly with drain wires at
both ends tied to ground and at 3m punch blocks cross-connected to the
110's (remember those, anyone?) within a 1/8 mile of a high-powered AM
broadcast facility. (Yeah, that was fun.)

--
Nate Duehr
nate@...

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Using maildrop with virtual users (MySQL db)

2008-09-28T15:09:29Z

Henry-Nicolas Tourneur ha scritto:

> Hello,
>
> I'v got some questions about the maildrop package. I don't know how to tell it
> to fetch
> users from the MySQL db (using courier-authdaemon).
>
> Here is my setup : Postfix (patched with VDA) / Courier(pop, imap and mysql) /
> Amavis / SpamAssassin and ... maildrop.
>
> The problem is that maildrop doesn't seems to use courier-authdaemon to check
> where it have to deliver email.
>
> The result is that I have a lot of messages like :
>
> Sep 28 16:12:01 ns203407 postfix/pipe[2368]: E2DE113B42:
> to=<henry.nicolas@...>, relay=maildrop, delay=0.03,
> delays=0.02/0/0/0.01, dsn=5.1.1, status=bounced (user unknown. Command
> output: Invalid user specified. )
>
> That make me think that he don't know where is the maildir for the user
> henry.nicolas@...
>
> So, I googled around for hours, without success and I'm writing this email
> with hope that somebody might have a solution.
>
> My final goals are :
> 1° Allow maildrop to act as my MDA, deliver spams to a junk folder is the home
> of the user.
>
> 2° Allow maildrop to manage my quotas (value stored in the MySQL db).
>
> Thanks in advance,
>
> Henry-Nicolas Tourneur.
>
>

could be a problem of permissions check :
- setuid bit of maildrop executable
- on authmysqlrc fields MYSQL_UID_FIELD and MYSQL_GID_FIELD
- UID and GID on the Maildir

Marco

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Using maildrop with virtual users (MySQL db)

2008-09-28T11:01:47Z

Henry-Nicolas Tourneur wrote:

Hi,

You can pickup all these configuration problems in the source code of
our control panel. For example, here's our /etc/courier/userdb
generation code:

http://git.gplhost.com/gitweb/?p=dtc.git;a=blob;f=admin/genfiles/gen_maildrop_userdb.php;h=99acab3b0a83752fa96eb331cf1fd9ee88f79b88;hb=3a2c6435db63ea0c6c9c8466bb2e5f91eb1a5008

You can also pickup all the code for doing all what you said above, our
panel has it all (and more).

Thomas

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-28T09:41:06Z

Hi,

"A. Dreyer" <adreyer@...> írta 2008-09-28 13:00-kor:

> > Second, its a bit hard to fix auto-negotiation in a vendors product such
> > as a switch, firewall or router. The sad fact is that not all hardware
> > plays nice with one another and you sometimes need to set the interfaces
> > to prevent problems. Yes that even means Gigabit interfaces. If the
> > interfaces negotiated correctly you would not configure the interface to
> > a set speed, only when they do not detect each other correctly. Maybe
> > that clears things up.
>
> If a device is sold as Gigabit-Switch/-Router/-Firewall and does not
> understand Gigabit auto-negotiation I would return it to the seller.

Yes! This is the correct attitude! I do not neither allow devices into my
network which doesn't respect standards. IEEE is not just a bunch of
idiots, and they don't do the standards just for enjoy themself.

<slightly-offtopic>
Common problem, when some moron come to here with bridging enabled in
his/her notebook, and connect to an access port, and then that alternative
OS, whichs name is not spoken - the evil one :-D - starts to send PDUs
towards the access port, and then it's change its state to err-disabled.
When they come to me for complain about the unreliable network, I tell
them, what they did: From my viewpoint, they accross the network policy, so
they can be happy, that I still allow them to connect, and re-allow the
err-disabled port.
</slightly-offtopic>

> During the specification of Gigabit Ethernet they tried to avoid the
> mistakes they made while defining 100Mbit/s and the auto-negotiation
> process in Gb/s is pretty clear - if any "Gigabit" device can not work
> with it I would expect other problems as well..
Yes! That's another correct observation! If a device doesn't comply a
standard which is so obvios, than what can do wrong too?
And the other observation is also correct: If you do too many hacks on your
network, it's usually undocumented, and it cause many other problems when
you connect devices which works correctly, and comply the standards.

PS.: Otherwise: I recently heard that, in the past few years, those "low
end manufacturers" like HP can create well-working devices which comply
standards ;-)

Summarized: If you make network with cheap devices, you pay it's price in
the TCO, with your many time you spent on debugging instead of operating,
and maybe the secondary consequences of the downtime is even higher if you
have to provide SLA or any kind of high availability.

Cheers,
gyu

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Using maildrop with virtual users (MySQL db)

2008-09-28T07:45:14Z

Hello,

I'v got some questions about the maildrop package. I don't know how to tell it
to fetch
users from the MySQL db (using courier-authdaemon).

Here is my setup : Postfix (patched with VDA) / Courier(pop, imap and mysql) /
Amavis / SpamAssassin and ... maildrop.

The problem is that maildrop doesn't seems to use courier-authdaemon to check
where it have to deliver email.

The result is that I have a lot of messages like :

Sep 28 16:12:01 ns203407 postfix/pipe[2368]: E2DE113B42:
to=<henry.nicolas@...>, relay=maildrop, delay=0.03,
delays=0.02/0/0/0.01, dsn=5.1.1, status=bounced (user unknown. Command
output: Invalid user specified. )

That make me think that he don't know where is the maildir for the user
henry.nicolas@...

So, I googled around for hours, without success and I'm writing this email
with hope that somebody might have a solution.

My final goals are :
1° Allow maildrop to act as my MDA, deliver spams to a junk folder is the home
of the user.

2° Allow maildrop to manage my quotas (value stored in the MySQL db).

Thanks in advance,

Henry-Nicolas Tourneur.

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-28T05:00:48Z

Shane Chrisp wrote:
> On Sun, 2008-09-28 at 12:25 +0100, ml10154@... wrote:
>> Shane Chrisp wrote:
>>> It is quite common practise to set the ports to what you need so that
>>> the auto-negotiation does not get it wrong, in particular setting half
>>> duplex instead of full duplex. This is not so common when its
>>> "cisco connected to cisco", but very common when its "cisco connected to
>>> some other brand".

>> Well it is common practise, but it is also one of the major problems.
>> People tend to forget these settings when they plug in other devices and
>> complain that they have "bad connectivity" or low throughput...
>>
>> Please make sure that you do not disable the auto-negotiation on
>> Gigabit-Ports (at least when running at 1Gbit/s) - it is marked as
>> mandatory in the standard and you should rather fix the broken part that
>> doesn't understand the auto-negotiation than fiddling around with these
>> settings.

> Firstly please don't send directly to me, i am subscribed to the list.
Sorry about that.

> Second, its a bit hard to fix auto-negotiation in a vendors product such
> as a switch, firewall or router. The sad fact is that not all hardware
> plays nice with one another and you sometimes need to set the interfaces
> to prevent problems. Yes that even means Gigabit interfaces. If the
> interfaces negotiated correctly you would not configure the interface to
> a set speed, only when they do not detect each other correctly. Maybe
> that clears things up.

If a device is sold as Gigabit-Switch/-Router/-Firewall and does not
understand Gigabit auto-negotiation I would return it to the seller.

During the specification of Gigabit Ethernet they tried to avoid the
mistakes they made while defining 100Mbit/s and the auto-negotiation
process in Gb/s is pretty clear - if any "Gigabit" device can not work
with it I would expect other problems as well..

Achim

--
Achim Dreyer || http://www.adreyer.com/
Internet Security Consultant || Phone: +44 7756948229
Senior Unix & Network Admin || RHCE, RHCA, CCNA, CCSA, CCSE, CSCE
CAcert Assurer || JNCIS-FW

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-28T04:42:47Z

On Sun, 2008-09-28 at 12:25 +0100, ml10154@... wrote:

> Shane Chrisp wrote:
> > It is quite common practise to set the ports to what you need so that
> > the auto-negotiation does not get it wrong, in particular setting half
> > duplex instead of full duplex. This is not so common when its
> > "cisco connected to cisco", but very common when its "cisco connected to
> > some other brand".
>
> Well it is common practise, but it is also one of the major problems.
> People tend to forget these settings when they plug in other devices and
> complain that they have "bad connectivity" or low throughput...
>
> Please make sure that you do not disable the auto-negotiation on
> Gigabit-Ports (at least when running at 1Gbit/s) - it is marked as
> mandatory in the standard and you should rather fix the broken part that
> doesn't understand the auto-negotiation than fiddling around with these
> settings.

Firstly please don't send directly to me, i am subscribed to the list.

Second, its a bit hard to fix auto-negotiation in a vendors product such
as a switch, firewall or router. The sad fact is that not all hardware
plays nice with one another and you sometimes need to set the interfaces
to prevent problems. Yes that even means Gigabit interfaces. If the
interfaces negotiated correctly you would not configure the interface to
a set speed, only when they do not detect each other correctly. Maybe
that clears things up.

Shane

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Partnership with debian.org

2008-09-28T04:28:40Z

Fellow Website Owner/Operator;

I am the marketing director of http://www.checkedyourcredit.com. We are a fairly
new website offering free credit reports and scores and although we are new we are
already receiving thousands of unique daily visitors do to our nationwide
advertising campaigns. I'm writing you personally to see if you would consider a
reciprocal link partnership with our website.

We are only accepting a limited number of link partners and our links page is only
one click away from our homepage which receives the majority of our traffic.
Because of this we ask that you place our link on a higher traffic area of your
website such as your homepage or the top of your links page.

If you would like to be included on our recommended links page please visit the
following page to choose from our various text and image ads:

http://www.checkedyourcredit.com/linkexchange.html

Please link to our website using one of the following text links or if you would
prefer you can select one of the banners or logo's from our website.

1) Credit Reports - Have you Checked Your Credit? Get your FREE credit report today!

2) Free Credit Report - Have you Checked Your Credit? Get your FREE credit report
today!

3) Credit Score - Have You Checked Your Credit? Get your FREE credit report and
score today!

Once you have added one of the above text links or banners to your website simply
send me an email and I will make sure to add your website on our next update. I
look forward to hearing from you. If you have any questons at all about our website
or our services feel free to contact me personally.

Sincerely;

Alicia Carasco - Internet Marketing Director
http://CheckedYourCredit.com

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-28T04:25:58Z

Shane Chrisp wrote:
> It is quite common practise to set the ports to what you need so that
> the auto-negotiation does not get it wrong, in particular setting half
> duplex instead of full duplex. This is not so common when its
> "cisco connected to cisco", but very common when its "cisco connected to
> some other brand".

Well it is common practise, but it is also one of the major problems.
People tend to forget these settings when they plug in other devices and
complain that they have "bad connectivity" or low throughput...

Please make sure that you do not disable the auto-negotiation on
Gigabit-Ports (at least when running at 1Gbit/s) - it is marked as
mandatory in the standard and you should rather fix the broken part that
doesn't understand the auto-negotiation than fiddling around with these
settings.

Achim

--
Achim Dreyer || http://www.adreyer.com/
Internet Security Consultant || Phone: +44 7756948229
Senior Unix & Network Admin || RHCE, RHCA, CCNA, CCSA, CCSE, CSCE
CAcert Assurer || JNCIS-FW

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-28T03:20:58Z

On Sun, 2008-09-28 at 10:33 +0200, PÁSZTOR György wrote:

> Hi,
>
> "Mr Shunz" <mrshunz@...> írta 2008-09-24 13:43-kor:
> > > Actually the switch is outside managed so i cannot myself make any changes.
> > > I do know it is set to 100 full. The switch does show errors on it also.
> > >
> >
> > cisco equipment likes fixed ports ... it *should* be enough to fix your
> > nics with ethtool like ethtool -s eth0 speed 100 duplex full autoneg off
> >
> Could you show us some proof for this, otherwise I have to assume, that
> this is a nonsense urban legend.
> I have tons of 2950s and 3750s at my workplace which is maintained by me,
> but there is no fixed ports in the whole building.

It is quite common practise to set the ports to what you need so that
the auto-negotiation does not get it wrong, in particular setting half
duplex instead of full duplex. This is not so common when its
"cisco connected to cisco", but very common when its "cisco connected to
some other brand".

Shane

smime.p7s (4K) Download Attachment

Re: load balanced nic

2008-09-28T01:33:06Z

Hi,

"Mr Shunz" <mrshunz@...> írta 2008-09-24 13:43-kor:
> > Actually the switch is outside managed so i cannot myself make any changes.
> > I do know it is set to 100 full. The switch does show errors on it also.
> >
>
> cisco equipment likes fixed ports ... it *should* be enough to fix your
> nics with ethtool like ethtool -s eth0 speed 100 duplex full autoneg off
>
Could you show us some proof for this, otherwise I have to assume, that
this is a nonsense urban legend.
I have tons of 2950s and 3750s at my workplace which is maintained by me,
but there is no fixed ports in the whole building.

Cheers,
gyu

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

kredit tanpa agunan 8-125 jt..adm/provisi 1,5 % saja

2008-09-27T22:21:35Z

PERSONAL LOAN HINGGA 200 jt,BUNGA 1,36-1,74% ..PERSYARATAN MUDAH...dr CITIBANK

2008-09-27T00:26:43Z

Nikmati pinjaman tanpa agunan/personal loan dengan persyaratan yang sangat mudah dan variatif

bunga yang kecil 1,36 % - 1,74 % dan proses hingga pencairan yang hanya membutuhkan waktu

maximal 5 hari kerja dengan persyaratan dokument yang sangat mudah .dr citibank

PERSYARATAN DOKUMENT PENGAJUAN (DENGAN MENGGUNAKAN CREDIT CARD )

1.FOTO COPY KTP

2.FOTO COPY CREDIT CARD

3.FOTO COPY BILLING 1 BULAN TERAKHIR

4.FOTO COPY NPWP ( UNTUK PINJAMAN DIATAS Rp.50 JT )

*MINIMAL PINJAMAN Rp.10.000.000*

*MAXIMAL PINJAMAN 3 X LIMIT CREDIT CARD ATAU Rp.100.000.000*

*BUNGA PINJAMAN Rp.10.000.000 - 24.000.000 = 1,74 % /BULAN*

*BUNGA PINJAMAN Rp.25.000.000 - 100.000.000 = 1,55 % /BULAN*

PERSYARATAN DOKUMENT PENGAJUAN (DENGAN MENGGUNAKAN SLIP GAJI )

1.FOTO COPY KTP

2.SLIP GAJI ASLI/SURAT KETERANGAN PENGHASILAN

3.FOTO COPY TRANSFER GAJI 3 BULAN TERAKHIR

4.FOTO COPY NPWP (PINJAMAN DIATAS 50 JT )

5.FOTO COPY CREDIT CARD

* MINIMAL GAJI PER BULAN RP.10.000.000,- *

* MINIMAL PINJAMAN Rp.40.000.000 *

* MAXIMAL PINJAMAN Rp.200.000.000 *

* BUNGA 1,36 % /BULAN *

hubungi :

JEFRI

021-94966624

or

jefri099@gmail.com

TABEL ANGSURAN

Nilai 12 bln 24 bln 36 bln

Pinjaman

10,000,000 1,007,330 590,670 451,780

11,000,000 1,108,063 649,737 496,958

12,000,000 1,208,796 708,804 542,136

13,000,000 1,309,529 767,871 587,314

14,000,000 1,410,262 826,938 632,492

15,000,000 1,510,995 886,005 677,670

16,000,000 1,611,728 945,072 722,848

17,000,000 1,712,461 1,004,139 768,026

18,000,000 1,813,194 1,063,206 813,204

19,000,000 1,913,927 1,122,273 858,382

20,000,000 2,014,660 1,181,340 903,560

21,000,000 2,115,393 1,240,407 948,738

22,000,000 2,216,126 1,299,474 993,916

23,000,000 2,316,859 1,358,541 1,039,094

24,000,000 2,417,592 1,417,608 1,084,272

25,000,000 2,470,825 1,429,175 1,081,950

26,000,000 2,569,658 1,486,342 1,125,228

27,000,000 2,668,491 1,543,509 1,168,506

28,000,000 2,767,324 1,600,676 1,211,784

29,000,000 2,866,157 1,657,843 1,255,062

30,000,000 2,964,990 1,715,010 1,298,340

31,000,000 3,063,823 1,772,177 1,341,618

32,000,000 3,162,656 1,829,344 1,384,896

33,000,000 3,261,489 1,886,511 1,428,174

35,000,000 3,459,155 2,000,845 1,514,730

40,000,000 3,953,320 2,286,680 1,731,120

45,000,000 4,447,485 2,572,515 1,947,510

50,000,000 4,941,650 2,858,350 2,163,900

55,000,000 5,435,815 3,144,185 2,380,290

60,000,000 5,929,980 3,430,020 2,596,680

65,000,000 6,424,145 3,715,855 2,813,070

70,000,000 6,918,310 4,001,690 3,029,460

75,000,000 7,412,475 4,287,525 3,245,850

80,000,000 7,906,640 4,573,360 3,462,240

85,000,000 8,400,805 4,859,195 3,678,630

90,000,000 8,894,970 5,145,030 3,895,020

95,000,000 9,389,135 5,430,865 4,111,410

100,000,000 9,883,300 5,716,700 4,327,800

hubungi :

JEFRI

021-94966624

or

"jefri099@gmail.com"

Re: ultramonkey / piranha - lvs solutions

2008-09-25T13:31:33Z

Ultramonkey and Heartbeat like parts of Linux-HA project.

--
"Al principio, el Hombre creó a Dios; y lo creó a su imagen y semejanza.
Y el Hombre dio a Dios multitud de nombres, y el poder de que fuera el Señor de toda la tierra cuando al Hombre le conviniere."

ultramonkey / piranha - lvs solutions

2008-09-24T16:34:23Z

Hi!

I have to deploy a veyr highly available www website with many servers
with virtualization.
I would prefer to use a gui to manage it (of course I like vim too ;) )
What solution do You advise and why ? :)

--
Wojciech Ziniewicz
Unix SEX :{look;gawk;find;sed;talk;grep;touch;finger;find;fl
ex;unzip;head;tail; mount;workbone;fsck;yes;gasp;fsck;more;yes;yes;eje
ct;umount;makeclean; zip;split;done;exit:xargs!!;)}

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: I need domain configuration exports from commercial control panels

2008-09-24T13:00:45Z

blorb wrote:
> i can export DA as backup.... would it help?

It would indeed! What it calls "backup" is the configuration of the
email accounts, etc. right?

Thomas

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: I need domain configuration exports from commercial control panels

2008-09-24T12:28:48Z

i can export DA as backup.... would it help?

On Wed, Sep 24, 2008 at 2:26 PM, Thomas Goirand <thomas@...> wrote:

Hi,

As you may know, I am the main author of the first open source control
panel that has a chance to be included in the forthcoming Lenny.

I wrote, 2 weeks ago, import and export functionalities to make it easy
to move domains between panel installations.

Now, I want to also be able to import from commercial control panels
like cPanel, DA or Plesk. I already have a (not surprisingly very ugly)
export from Plesk 8, but I need some from others.

So if you can help me, fill a fake domain with accounts (email,
subdomains, ftp, database), then do an export, and send it to me. That
would be of great help.

Regards and thanks to anybody that can help,

Thomas Goirand

P.S: Note that I tried to ask in other places without much success...

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

--
"I thought what I'd do was, I'd pretend I was one of those deaf-mutes".
echo tibibsgAhnbjm/dpn | perl -pe 's/(.)/chr(ord($1)-1)/ge'

Re: some weird upgrade problem

2008-09-24T11:49:47Z

On Sun, Sep 07, 2008 at 02:53:00AM +0100, Paweł Krzywicki wrote:

> Hi,
> I have tried upgrade my Debian yesterday and apt showed me an weird error.
> I am having the problem with python-libxml2 package. I have also
> in /usr/share/doc/python-libxml2/examples directory files like that:
>
> ?????????? ? ? ? ? ? schema.py
> ?????????? ? ? ? ? ? serialize.py
> ?????????? ? ? ? ? ? sync.py
>
> date of creation is of these files is 1st Jun1970 ?!
> I cannot do absolutely nothing with them. I have tried to use live CD and
> mount my drive not as a system one and remove them (rm -f), change owner of
> these files, nothing seems to be working with them. Every time when I am
> trying to do something I have straight something like: "You don't have an
> access to the file" (?!) Do you have any ideas how to fix that.
> thanks in advance

It's likely that it's just the lack of execute permission on the containing
directory:

mowsiany@beczulka:~$ sudo mkdir a
mowsiany@beczulka:~$ sudo touch a/a
mowsiany@beczulka:~$ ls -al b
total 8
drwxr-xr-x 2 root root 4096 2008-09-24 19:47 .
drwxr-xr-x 74 mowsiany eng 4096 2008-09-24 19:47 ..
-rw-r--r-- 1 root root 0 2008-09-24 19:47 a
mowsiany@beczulka:~$ sudo chmod a-x a
mowsiany@beczulka:~$ ls -al a
total 0
?--------- ? ? ? ? ? a/.
?--------- ? ? ? ? ? a/..
?--------- ? ? ? ? ? a/b
mowsiany@beczulka:~$

--
Marcin Owsiany <porridge@...> http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

RE: Multiple virtual interfaces in one VLAN

2008-09-24T07:46:42Z

I've had great success with the following in /etc/network/interfaces:

iface eth0.400:0 inet static
...

iface eth0.400:1 inet static
...

iface eth0.400:2 inet static
...

And so on. Using to manage wireless CPEs in different subnets on the
same VLAN.

Are you trying to make all of your aliases use DHCP or statics? I would
expect multiple "use DHCP" entries on the same NIC would take you
nowhere - if work at all.

Cheers
---
Ross Halliday
Network Operations
WTC Communications

-----Original Message-----
From: Michael Fischer [mailto:michi.fischer@...]
Sent: Wednesday, September 24, 2008 9:17 AM
To: debian-isp@...
Subject: Multiple virtual interfaces in one VLAN

hi,

i have a machine with only one physical network interface.

i know that it is possible to create virtual interfaces (eth0:1, etc.)
to have multiple ip-adresses on one physical interface.

now i would need to configure vlans on this machine - no problem so far,
this is my config:

[code]
auto eth1.1001
iface eth1.1001 inet dhcp

auto eth1.1012
iface eth1.1012 inet dhcp
[/code]

that works too, i get the interfaces vlan1012 and vlan1001 and ip
adresses in the corresponding subnets.

but i cant create virtual interfaces for those vlan interfaces - the
problem is i would need about 10 interfaces in one vlan.

i have tried both eth1.1001:1 and eth1:1.1001 syntax - both dont work.

does anyone know how to do this?

lg, Michi
--
email: michi.fischer@...
web: http://www.webfischer.at

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact
listmaster@...

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Multiple virtual interfaces in one VLAN

2008-09-24T06:25:50Z

Michael Fischer wrote:
> i cant create virtual interfaces for those vlan interfaces - the
> problem is i would need about 10 interfaces in one vlan.
>
> i have tried both eth1.1001:1 and eth1:1.1001 syntax - both dont work.
>
> does anyone know how to do this?

Don't use aliases, use iproute2 to add the extra IP addresses to the
interfaces:

ip address add 192.168.5.5/24 dev eth1.1001

Regards,

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Multiple virtual interfaces in one VLAN

2008-09-24T06:16:31Z

hi,

i have a machine with only one physical network interface.

i know that it is possible to create virtual interfaces (eth0:1, etc.)
to have multiple ip-adresses on one physical interface.

now i would need to configure vlans on this machine - no problem so far,
this is my config:

[code]
auto eth1.1001
iface eth1.1001 inet dhcp

auto eth1.1012
iface eth1.1012 inet dhcp
[/code]

that works too, i get the interfaces vlan1012 and vlan1001 and ip
adresses in the corresponding subnets.

but i cant create virtual interfaces for those vlan interfaces - the
problem is i would need about 10 interfaces in one vlan.

i have tried both eth1.1001:1 and eth1:1.1001 syntax - both dont work.

does anyone know how to do this?

lg, Michi
--
email: michi.fischer@...
web: http://www.webfischer.at

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: load balanced nic

2008-09-24T04:43:51Z

> Actually the switch is outside managed so i cannot myself make any changes.
> I do know it is set to 100 full. The switch does show errors on it also.
>

cisco equipment likes fixed ports ... it *should* be enough to fix your
nics with ethtool like ethtool -s eth0 speed 100 duplex full autoneg off

cheers

--
------------------------------------------------
Daniele Santi .o.
daniele@... ..o () ascii ribbon campaign
Linux User #415108 ooo /\ www.asciiribbon.org
------------------------------------------------

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

I need domain configuration exports from commercial control panels

2008-09-24T04:26:26Z

Hi,

As you may know, I am the main author of the first open source control
panel that has a chance to be included in the forthcoming Lenny.

I wrote, 2 weeks ago, import and export functionalities to make it easy
to move domains between panel installations.

Now, I want to also be able to import from commercial control panels
like cPanel, DA or Plesk. I already have a (not surprisingly very ugly)
export from Plesk 8, but I need some from others.

So if you can help me, fill a fake domain with accounts (email,
subdomains, ftp, database), then do an export, and send it to me. That
would be of great help.

Regards and thanks to anybody that can help,

Thomas Goirand

P.S: Note that I tried to ask in other places without much success...

--
To UNSUBSCRIBE, email to debian-isp-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...