| |
| |
| |
|
Page: 1 2 3 4 5
Comments:
<0> yashy: what about you, mate?
<0> if u have issue, then get a box of tissue,
<0> next
<1> thanks!
<0> :)
<0> u can always get a razor and dice your .....
<0> oops, Dan brown lost his case. movie wont be coming out now
<2> not my day... definitely not my day.
<3> anyone rocking postfix+mysql+maildir?
<1> just taking a random poll?
<3> nah
<3> I have a question
<3> trying to figure out how to get postfix to deliver emails to the maildir specified in the DB in maildir format
<4> i thought postfix was an mta
<3> and?
<4> an mta has no concept of local mailbox formats
<4> since it doesn't deliver to them
<1> it's the MDA that delivers the mail
<1> Like courier
<4> that's the job of a local delivery agent
<1> I manually changed nothing in the last 24 hours, and nat just stopped working this aft :/
<3> I can get postfix to deliver the email in Maildir format to $home/Maildir
<3> but I want it to lookup at the DB to verify the user homedir
<1> 18:23:53.516017 IP mail.yashy.com > 0.0.0.0: pfsync 228
<1> 18:23:53.702177 IP mail.yashy.com > 0.0.0.0: pfsync 228
<1> are these normal packets to see in tcpdump?
<5> http://s2.bitefight.pl/c.php?uid=38393
<2> one channel to late for the kill.
<6> Hi there, means "cached" in this line: Swap: 8703968k av, 0k used, 8703968k free 3422624k cached ?
<6> What means
<7> probably refers to filesystem data that's currently held in ram
<6> 3 GB held in RAM ?
<6> What happens if the system crash then? some important informations could be lost...
<7> doesn't work that way, for the most part
<7> how much ram is in the system?
<6> 24 GB
<7> so 3GB is completely plausible
<2> I'm still surprised that Overhaulin' hasn't been sued out of existance.
<7> in general, a filesystem cache will store data in the expectation of a second read, instead of to delay or defer a write
<6> yes, but why those filesystem data are not saved into the HDD?
<7> ie, it mostly does things that are pretty safe
<2> jb, they are on the harddrive.
<2> Just they are stashed in ram for faster access.
<7> you can experiment with this..
<6> oh ok, I see
<7> pick a drive, and create a 128MB file on it
<7> then run: time dd if=file bs=64k of=/dev/null
<7> and then run it again
<7> of course, depending how the caching works, it may be entered into the cache on creation, in which case the test may be ambiguous
<7> but you may also see that the second time reading the file is much faster than the first
<6> Yes I understand now... It's a bit like the cache memory you have inside some HDD ( for exemple 8 or 16 MB)
<7> better than that, usually.. it doesn't have to go through a relatively slow m*** storage protocol / cable..
<7> filesystems do sometimes do write caching too, however.. for instance to reorder writes so that they can be written contiguously
<6> but, why this info appears on the "Swap" line
<7> why not?
<6> because "Swap" is using HDD like a RAM
<2> jb23, I suspect because Linux's cache is malluable. If that ram is needed for real applications it will give it up. So its only an illusion the ram is used.
<2> just like swap is an illusion of real memory.
<6> Cache is using RAM like a HDD
<7> again, it's common to use free ram for filesystem buffering
<7> if some big process needs the ram, an os will typically dump the cache info to free some up
<2> malleable
<2> whatever.
<7> it costs almost nothing to do that, so it's a common optimisation
<6> ok, thank you very much for the explanation :)
<2> 9 out of 10 times they don't detect that they've already connected with a "human"
<2> And thus there is no way to tell them no to call you anymore.
<2> /no/not/
<8> im having problems p***ing alt-key strokes through an xterm to an app. I think its xterms fault.. dosent happen from console or via putty
<7> try switching your "num lock" key setting if your keyboard has one
<7> (some window managers use this to determine which key sequences they should trap themselves, and which should be p***ed to applications)
<1> okay.. who wants to help me with my pf nat issue? :)
<1> I'm seeing packets outbound only on my 192.168. (rl0) but nothing from the client machines on the external interface (hme0/tun0)
<1> rl0 = internal lan interface
<9> can anyone look at my ipfw script to see why its blocking everything?
<1> 21:56:17.750229 PPPoE [ses 0x7f8] IP mail.yashy.com > 0.0.0.0: pfsync 228
<1> 21:56:18.805201 PPPoE [ses 0x7f8] IP mail.yashy.com > 0.0.0.0: pfsync 228
<1> I'm also seeeing a flood of these in tcpdump, are they normal? :)
<10> yup, its Mouring trying to hack your server
<7> you're sending pfsync frames out a non private interface?
<1> only the client machines behind it don't seem to be able to get through it for some reason
<9> http://juggalo.be:81/ipfw.sh anyone know if this is too strict?
<1> Figz: not intentionally :) Doesn't look like they're going anywhere anyways. Should there be a pf.conf rule for pfsync?
<7> yashy, do you have a hostname.pfsync*?
<1> # ls -la /etc/*pfsync*
<1> ls: No match.
<7> it's possible whatever's happening on that interface just has the same protocol number as pfsync, and isn't related
<7> ifconfig pfsync0
<7> is it up?
<1> pfsync0: flags=0<> mtu 2020
<1> that's it
<7> nope.. so the stuff is presumably coming from somewhere other than pfsync..
<1> pflog0: flags=141<UP,RUNNING,PROMISC> mtu 33160
<7> not related
<1> other than net.inet.ip.forwarding: 1
<7> pfsync is a protocol for synchronising pf sate information between redundant routers
<7> (usually in conjunction with carp)
<1> I'm not sure how to get the rl0 packets to go out on tun0 other than the nat entry
<7> kernel routes based on destination address
<1> I have no need for pfsync then.. I should turn it off.
<7> if tun0 bears the default gateway, all packets not directed by some more specific destination will go there automatically
<1> $ telnet fark.com 80
<1> Trying 207.58.150.100...
<1> I try this from a client machine and it just hangs
<1> I see the packets hit rl0, but get no replies.
<1> And nothing out or in on tun0/hme0
<7> can the client ping the pf gateway?
<1> yes
<1> I'm ssh'd to it right now to IRC
<1> I just can't go "through it"
<7> is pf enabled?
<1> Status: Enabled for 0 days 07:27:55 Debug: Urgent
<7> is ip forwarding set to 1?
<1> I did pfctl -Fr (flush rules, so I only have the nat entry) and still no luck
<1> # sysctl -a | grep forwarding
<1> net.inet.ip.forwarding: 1
<7> must be freebsd..
<7> show the output of: pfctl -snat
<1> # pfctl -sn
<1> nat on tun0 inet from 192.168.1.0/28 to any -> 206.248.137.44
<1> rdr inet proto tcp from <spamd> to any port = smtp -> 127.0.0.1 port 8025
<7> what's the output of: ifconfig tun0
<1> # ifconfig tun0
<1> tun0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1492
<1> inet 206.248.137.44 --> 65.39.134.4 netmask 0xffffffff
<1> Opened by PID 808
<7> why do you use such a narrow mask (/28) on 192.168.1.0?
<1> it's only my home lan
<7> is the client's IP between 192.168.1.1 and 192.168.1.14?
<1> I have less than 15 computers left in my house, and 10 of those are for sale
<1> .8 in this case
<9> what perm should /etc/master.p***wd be?
<7> smokey, 700, root.wheel
<1> dhcpd only gives out between 2 and 14
<9> wait, let me rephrase, what is the best perm for /etc period?
<2> -rw------- 1 root wheel 2439 Nov 15 16:37 /etc/master.p***wd
<7> err, 600 root.wheel
<9> should i set all files to 600 ?
<2> no
<7> don't be stupid
<9> because i have some poor hacker trying to access /etc files.
Return to
#unixhelp
or
Go to some related
logs:
#computers
python subscriptable type
trigger regex parser
#online
xls to cvs
#unixhelp
klute mosquito festival
#politics
#tcl
#firebird
|
|