Re: create a backup of an online server
On Wed, 28 Dec 2011 10:06:14 +0400 "Wesley M." wrote: > Hi, > I want to backup our mailserver(4.7) in production. > I read : > http://www.openbsd.org/faq/faq10.html#DupFS > > Can i do this wd1(my backup > disk) : > > mount /dev/wd1a /mnt > dump -0auf /mnt/etc_backup /dev/wd0a > ... > same > for wd0d and wd0e ... Or do i need absolutely to do it in Single > User? > > Or > perhaps, there's a better way to do it. Personaly I found a way for myself to produce a consistent backups: 1. have an additional filesystem /backup 2. mount /backup 3. sync your changed files to /backup (I usually do this with rsync, YMMV) 4. umount /backup 5. mount /your_backup_media 6. dump -0auf /your_backup_media /dev/ However it need some preparation done beforehand, preferably on system install. this way you will be able to dump live filesystems without having to reboot your server. -- With best regards, Gregory Edigarov
file name too long error - apache
Hello list, I'm trying to install an PHP web app that generates very long URLs on my OpenBSD 5.0 i386 snapshot. But I just keep getting "file name too long" errors in apache error log like below. It happened on both built-in apache 1.3 and apache-httpd (2.2.20) in ports. I even tried compile apache2 using source files from ubuntu, as the same app works fine on ubuntu 11.04 + apache2 (2.2.17), it didn't help either. Any insight or suggestions? [Wed Dec 28 14:15:51 2011] [error] [client 10.249.70.50] (63)File name too long: Cannot map GET /media/ajax/component/boxList/filter/featured/limit/all/layout/thumbBig/vars/ a%253A25%253A%257Bs%253A3%253A%2522act%2522%253Bs%253A7%253A%2522boxList%2522 %253Bs%253A3%253A%2522mod%2522%253Bs%253A5%253A%2522media%2522%253Bs%253A4%25 3A%2522mode%2522%253Bs%253A3%253A%2522all%2522%253Bs%253A6%253A%2522filter%25 22%253Bs%253A8%253A%2522featured%2522%253Bs%253A5%253A%2522limit%2522%253Bs%2 53A3%253A%2522all%2522%253Bs%253A6%253A%2522layout%2522%253Bs%253A8%253A%2522 thumbBig%2522%253Bs%253A6%253A%2522search%2522%253Bs%253A0%253A%2522%2522%253 Bs%253A8%253A%2522per_page%2522%253Ba%253A3%253A%257Bs%253A8%253A%2522thumbBi g%2522%253Bi%253A6%253Bs%253A5%253A%2522thumb%2522%253Bi%253A10%253Bs%253A4%2 53A%2522list%2522%253Bi%253A4%253B%257Ds%253A11%253A%2522show_filter%2522%253 Bb%253A1%253Bs%253A10%253A%2522show_limit%2522%253Bb%253A0%253Bs%253A11%253A% 2522show_layout%2522%253Bb%253A1%253Bs%253A11%253A%2522show_search%2522%253Bb %253A0%253Bs%253A10%253A%2522show_pager%2522%253Bb%253A0%253Bs%253A9%253A%252 2show_more%2522%253Bb%253A1%253Bs%253A9%253A%2522save_page%2522%253Bb%253A1%2 53Bs%253A10%253A%2522pager_name%2522%253Bs%253A4%253A%2522page%2522%253Bs%253 A9%253A%2522thumbsize%2522%253Bs%253A7%253A%2522160x120%2522%253Bs%253A9%253A %2522more_link%2522%253Bs%253A10%253A%2522media%252Flist%2522%253Bs%253A2%253 A%2522id%2522%253Bs%253A9%253A%2522media-box%2522%253Bs%253A9%253A%2522compon ent%2522%253Bs%253A7%253A%2522boxList%2522%253Bs%253A4%253A%2522type%2522%253 BN%253Bs%253A4%253A%2522text%2522%253BN%253Bs%253A13%253A%2522captionParams%2 522%253Ba%253A0%253A%257B%257Ds%253A7%253A%2522caption%2522%253Bs%253A9%253A% 2522New%2Bmedia%2522%253Bs%253A4%253A%2522page%2522%253Bi%253A1%253B%257D HTTP/1.1 to file, referer: http://host BTW, I found this url https://issues.apache.org/bugzilla/show_bug.cgi?id=45187 while googling, it did help on understanding the possible root cause, but did not help in resolving it. thanks, Alan Cheng
Re: ufs journal ?
On Tue, Dec 27, 2011 at 01:17:26PM -0800, Chris Cappuccio wrote: > Otto Moerbeek [o...@drijf.net] wrote: > > > > There are several way to speedup fsck which are available now: > > > > - Use larger block and fragment sizes when doing a newfs, of course > > this requires rebuilding the file system > > Is there any sort of rule-of-thumb for this now that 1TB drives are "cheap" > and 2 and 3TB drives are available? In arrays, no less. I'd say the bigger the better, (max for both is 64k), unless you expect to store a lot of small files. In that case having large blocks and fragments will spoil a lot of disk space and your number of inodes might be too small. -Otto
create a backup of an online server
Hi, I want to backup our mailserver(4.7) in production. I read : http://www.openbsd.org/faq/faq10.html#DupFS Can i do this wd1(my backup disk) : mount /dev/wd1a /mnt dump -0auf /mnt/etc_backup /dev/wd0a ... same for wd0d and wd0e ... Or do i need absolutely to do it in Single User? Or perhaps, there's a better way to do it. Thank you very much. Wesley.
Where do I buy Lemote Loongson/Godson MIPS hardware? (was Re: Longsoon/Godson MIPS boxes, where to buy?)
On Tue, Dec 27, 2011 at 10:09 AM, Dave U. Random wrote: > Are the Longson/Godson MIPS boxes available over the counter yet? If so > where is the best place to order one? Thanks. A brief search of the archives gives a few resources. Spelling the architecture right helps, but searching for "lemote" does wonders. Start reading here: http://openbsd.org/loongson.html Relevant threads on misc@: http://marc.info/?l=openbsd-misc&w=2&r=1&s=lemote&q=b http://marc.info/?l=openbsd-misc&w=2&r=1&s=loongson&q=b Acquiring hardware: International: http://www.aliexpress.com/wholesale?SearchText=loongson&catId=0 http://www.aliexpress.com/wholesale?SearchText=lemote&catId=0 In China (drop shipment to a forwarder may be necessary): http://loogson.taobao.com/ In Europe: http://www.tekmote.nl In the US, Amazon has a direct sales from Freedom Included. Prime eligible. http://www.amazon.com/s?ie=UTF8&search-alias=computers&field-manufacturer=Lemote http://freedomincluded.com/ Compiled for the archives, YMMV.. Special thanks to Miod, Diana and others for their postings.
Re: Longsoon/Godson MIPS boxes, where to buy?
On Wed, Dec 28, 2011 at 2:09 AM, Dave U. Random < anonym...@anonymitaet-im-inter.net> wrote: > Are the Longson/Godson MIPS boxes available over the counter yet? If so > where is the best place to order one? Thanks. > > checkout http://www.tekmote.nl/
Re: What is wrong with this pf config
> I wish they would ban you from this list already. I'm sick of seeing > your reply to every thread when you never have anything constructive to > say. Then killfile the stupid sonofabitch like the rest of us and stop quoting him. Then we won't have to see his posts. > > > To our sweethearts and wives. May they never meet. -- 19th century > toast To Tate's feeble skull and a baseball bat. May they meet soon!
Longsoon/Godson MIPS boxes, where to buy?
Are the Longson/Godson MIPS boxes available over the counter yet? If so where is the best place to order one? Thanks.
Re: ufs journal ?
Otto Moerbeek [o...@drijf.net] wrote: > > There are several way to speedup fsck which are available now: > > - Use larger block and fragment sizes when doing a newfs, of course > this requires rebuilding the file system Is there any sort of rule-of-thumb for this now that 1TB drives are "cheap" and 2 and 3TB drives are available? In arrays, no less.
Re: CF Card setup
pe...@bsdly.net (Peter N. M. Hansteen) writes: > for a simple dhcp setup, or for a fixed address and a specific link > speed something like (lifted from man hostname.if) > > inet 10.0.0.1 255.255.255.0 10.0.0.255 description "Bob's uplink" actually that does not specify a line speed, but the man pages will tell you how to do that too :) -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/ "Remember to set the evil bit on all malicious network traffic" delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
Re: CF Card setup
Jannik Pruitt writes: > ifconfig says > lo0 (up loop back running multicast mtu 33196 > R10 Media enthernet auto (This is normally the IP address I think but > there is not one) > Enco - also active > PFlog0 also active I assume "R10" is actually rl0, indicating that the Ethernet card is a Realtek based part (see man rl). You need to put together a valid config file for that one, /etc/hostname.rl0, with appropriate contents using your favorite text editor (mg and vi are in the base system). Reading at least man hostname.if and http://www.openbsd.org/faq/faq6.html is a really good idea at this point. Most likely you need a file that consists of a single line, either dhcp for a simple dhcp setup, or for a fixed address and a specific link speed something like (lifted from man hostname.if) inet 10.0.0.1 255.255.255.0 10.0.0.255 description "Bob's uplink" in which case you will also need to add useful content to /etc/resolv.conf (and likely /etc/hosts) for name resolution to work plus possibly a few other wrinkles such as enabling forwarding if it's a gateway you're building, and so on. The best place to start is to read the relevant parts of the FAQ and the man pages. OpenBSD documentation is both accessible and useful, and if you're still stuck some of us have written supplementary docs that are not that hard to find. Or come back here, reasonable questions usually generate somewhat useful answers. -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/ "Remember to set the evil bit on all malicious network traffic" delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
Re: CF Card setup
Hi Nick and Joakim thanks for writing back. I am not really sure of the rules. This is the first time I have ever used any BSD except for an Apple PC. I do have some linux experience. I was running Puppy Linux on my old Netier XL1000 box and it was not doing so well even with puppy and I wanted to learn more about BSD. I checked the basic system requirements and I have 256mb an S3 graphics card a common Tbase T card built on the board. I assume the ifconfig is about the same as linux. The networking worked well with Puppy. The end goal is have a small file server, web server, emacs, sc, bc and alpine on the little box. It has 32GB CF card. I thought about slackware also as it has CF install built in but again I heard so much from the podscast I wanted to give it try. So this is what happens when I type in www.google.com host not reachable, my router also, ssh sdf.lonestar.org also not working ifconfig says lo0 (up loop back running multicast mtu 33196 R10 Media enthernet auto (This is normally the IP address I think but there is not one) Enco - also active PFlog0 also active When I log in it says: openBSD/i386 (Netier1000.speedport_W_723V_type_A) (TTYC0) >From the answers I got I think i have change the Enternet card setting. But there is no command to run the networking or Xgraphics set up again? On Wed, Dec 21, 2011 at 10:01 AM, Joakim Aronius wrote: > * Jannik Pruitt (pruttel...@googlemail.com) wrote: > > Is there a way to make the setup come up again? > > If there is a RS232 serial port you could use that to get console access > using a serial cable from som other machine. > http://www.openbsd.org/faq/faq7.html#SerCon > > This is really good to have if you do not have a graphics card in the box. > Not only for initial install but also for upgrades to new releases! You can > find cheapo USB-serial converters on eBay etc if you don't have RS232 on > e.g. the laptop you want to connect to the thin client. > > /joakim
pf state key linking mismatch
Hi, I can't really find anything explaning these error except that some said that you never want it to happen and Henning writing that it could be ignore in some case. In my case, I think I should ignore it but would like to understand it just to be sure. pf: state key linking mismatch! dir=OUT, if=em1, stored af=2, a0: 0.0.0.0:53504, a1: 68.67.XXX.XXX:47406, proto=17, found af=2, a0: 68.67.XXX.XXX:47406, a1: 10.8.0.1:53504, proto=17 68.67.XXX.XXX is inside my network and the're not any nat rules for this IP adress since it's a public one. em1 is my internal if with ip 10.8.0.1. Proto 17 = udp a0: 0.0.0.0:53504 ??? What could this IP address meen ? For sure, it's not a broadcast. Which of those are src and dst ip address ? My nat rules : match on $ext_if from 10.8.0.2 binat-to 64.119.XXX.XXX match out on $ext_if inet from nat-to $ext_if pfctl -t ipnat -Ts 10.0.0.0/8 96.XXX.XXX.0/20 216.XXX.XXX.0/24 216.XXX.XXX.0/24 216.XXX.XXX.0/24 216.XXX.XXX.0/24 216.XXX.XXX.0/24 Those public IP address are only in the NAT rules because the should never goes out by this edge except in case big problem at our other edge that have some fibers in redondancy. It will be replace by BGP eventually. Thanks Michel
Re: ftpd change?
There were some DoS protections added which limit the number of results that can be returned from a glob, it is to do with total filename lengths iirc. I see this a lot in package snapshot directories. If you want something good that will handle more files, try pureftpd (it's in ports). On 2011-12-26, STeve Andre' wrote: > I have an old FTP machine (3.6!) that a horde of people are > suckling from at the moment. I'm making a new one remotely, > but found that an 'mget *' on the new machine fails, with no > such file. There are 515 files in the dir; the 3.6 system expands > the mget, and the newer 4.9-current doesn't. > > I don't see anything in the ftpd man page about this. Was > this a change, or did I miss something? I've currently got 110 > people feeding from a 128M machine and would like to put > this larger machine into service right now. ;-) > > Thanks! > > --STeve Andre'