Re: sftp and shell access
On Dec 14, 2004, at 02:11, Josh Paetzel wrote: I am looking for a way to give a user an sftp account without giving them a shell. So far I've tried setting their shell to /sbin/nologin, but when they try to log in via sftp it gives them a "message to long" error. Any pointers would be appreciated...I've tried the FAQ, handbook and google so far. sftp uses a ssh connection to tunnel to ftp. The connection is actually made to your ssh port. There is also ftps which is ftp with ssh imbedded in it (like https). With that the connection is actually made to fhe ftp server port. ftps is available in the ports (BSDftpd-ssl). Since it doesn't use ssh you can set the user to not have login capability. Clients for ftps or sftp are not always easy to find. The web page for BSDftpd-ssl does list a number of compatable clients that are available. I suspect that sometime there will be a general shift to one of those approaches and the other will go away which would make it easier to find clients. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
5.3 Building Kernel/World
The instructions for building world/kernel for 4.x are straight forward and work fine. However, I seem to have munged two 5.3 installations now. I have been through all the UPDATING notes and the handbook and something is obviously not clear. The approach I used is: Clean install from 5.3 distribution. make buildworld create new config file LAFN make buildkernel KERNCONF=LAFN make installkernel KERNCONF=LAFN reboot make installworld reboot At that point It appeared I was using the LAFN kernel rather than generic. However, tonight I tried to make a new kernel. NO go: ERROR: version of config(8) does not match kernel! config version = 500012, version required = 500013 So I tried to reinstall the kernel: make installkernel KERNCONF=LAFN install -o root -g wheel -m 555 acpi.ko /boot/kernel install: acpi.ko: No such file or directory How are you supposed to build a new kernel that works? How do I recover this? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: 5.3 Building Kernel/World
On Jan 16, 2005, at 22:05, Kris Kennaway wrote: On Sun, Jan 16, 2005 at 09:51:13PM -0800, Doug Hardie wrote: The instructions for building world/kernel for 4.x are straight forward and work fine. However, I seem to have munged two 5.3 installations now. I have been through all the UPDATING notes and the handbook and something is obviously not clear. The approach I used is: Clean install from 5.3 distribution. make buildworld create new config file LAFN make buildkernel KERNCONF=LAFN make installkernel KERNCONF=LAFN reboot make installworld reboot At that point It appeared I was using the LAFN kernel rather than generic. However, tonight I tried to make a new kernel. NO go: ERROR: version of config(8) does not match kernel! config version = 500012, version required = 500013 So I tried to reinstall the kernel: make installkernel KERNCONF=LAFN install -o root -g wheel -m 555 acpi.ko /boot/kernel install: acpi.ko: No such file or directory How are you supposed to build a new kernel that works? How do I recover this? The only way I can think for this to happen is if your source tree was inconsistent (i.e. not completely updated), or you updated your sources after you did the installworld, and the kernel depends on the newer version of config than the one you have built (in this case the 'safe' buildworld/buildkernel/installkernel method you used to begin with should still work). Kris those instructions were typed just as noted with nothing inbetween. I have replaced kernel with kernel.old so the system will boot. But now I have an old kernel and new world (possibly). Nothing for reconstruction seems to work. buildkernel continues to give the above error. I guess I'll try a buildworld again tomorrow. Don't know what else to do. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: 5.3 Building Kernel/World
On Jan 16, 2005, at 22:05, Kris Kennaway wrote: On Sun, Jan 16, 2005 at 09:51:13PM -0800, Doug Hardie wrote: The instructions for building world/kernel for 4.x are straight forward and work fine. However, I seem to have munged two 5.3 installations now. I have been through all the UPDATING notes and the handbook and something is obviously not clear. The approach I used is: Clean install from 5.3 distribution. make buildworld create new config file LAFN make buildkernel KERNCONF=LAFN make installkernel KERNCONF=LAFN reboot make installworld reboot At that point It appeared I was using the LAFN kernel rather than generic. However, tonight I tried to make a new kernel. NO go: ERROR: version of config(8) does not match kernel! config version = 500012, version required = 500013 So I tried to reinstall the kernel: make installkernel KERNCONF=LAFN install -o root -g wheel -m 555 acpi.ko /boot/kernel install: acpi.ko: No such file or directory How are you supposed to build a new kernel that works? How do I recover this? The only way I can think for this to happen is if your source tree was inconsistent (i.e. not completely updated), or you updated your sources after you did the installworld, and the kernel depends on the newer version of config than the one you have built (in this case the 'safe' buildworld/buildkernel/installkernel method you used to begin with should still work). Kris Well, I tried makeworld again. Dies in Step 3. Reloaded all source from the distribution CD. makeworld dies in exactly the same place: ===> gnu/usr.bin/binutils/libbfd cc -O -pipe -I. -I/usr/src/gnu/usr.bin/binutils/libbfd/i386 -I/usr/src/gnu/usr.bin/binutils/libbfd -I/usr/src/gnu/usr.bin/binutils/libbfd/../libbfd -I/usr/obj/usr/src/i386/usr/src/gnu/usr.bin/binutils/libbfd/../libbfd -I/usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/ include -D_GNU_SOURCE -I/usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd -DSELECT_ARCHITECTURES=" &bfd_i386_arch" -DHAVE_bfd_elf32_i386_freebsd_vec -DHAVE_bfd_elf32_i386_vec -DSELECT_VECS=" &bfd_elf32_i386_freebsd_vec ,&bfd_elf32_i386_vec" -DDEFAULT_VECTOR=bfd_elf32_i386_freebsd_vec -I/usr/obj/usr/src/i386/legacy/usr/include -c /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ cpu-i386.c In file included from /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ cpu-i386.c:23: /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:114: error: syntax error before "_bfd_add_bfd_to_archive_cache" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:115: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:116: error: syntax error before "_bfd_generic_mkarchive" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:117: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:120: error: syntax error before "bfd_slurp_armap" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:121: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:122: error: syntax error before "bfd_slurp_bsd_armap_f2" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:123: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:126: error: syntax error before "_bfd_slurp_extended_name_table" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:127: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:128: error: syntax error before "_bfd_construct_extended_name_table" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:129: error: syntax error before "bfd_boolean" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:129: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:130: error: syntax error before "_bfd_write_archive_contents" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:131: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:132: error: syntax error before "_bfd_compute_and_write_armap" /usr/src/gnu/usr.bin/binutils/libbfd/../../../../contrib/binutils/bfd/ libbfd.h:133: warning: data definition has no type or storage class /usr/src/gnu/usr.bin/binut
Re: 5.3 Building Kernel/World
On Jan 16, 2005, at 23:19, Kris Kennaway wrote: On Sun, Jan 16, 2005 at 11:15:23PM -0800, Doug Hardie wrote: On Jan 16, 2005, at 22:05, Kris Kennaway wrote: On Sun, Jan 16, 2005 at 09:51:13PM -0800, Doug Hardie wrote: The instructions for building world/kernel for 4.x are straight forward and work fine. However, I seem to have munged two 5.3 installations now. I have been through all the UPDATING notes and the handbook and something is obviously not clear. The approach I used is: Clean install from 5.3 distribution. make buildworld create new config file LAFN make buildkernel KERNCONF=LAFN make installkernel KERNCONF=LAFN reboot make installworld reboot At that point It appeared I was using the LAFN kernel rather than generic. However, tonight I tried to make a new kernel. NO go: ERROR: version of config(8) does not match kernel! config version = 500012, version required = 500013 So I tried to reinstall the kernel: make installkernel KERNCONF=LAFN install -o root -g wheel -m 555 acpi.ko /boot/kernel install: acpi.ko: No such file or directory How are you supposed to build a new kernel that works? How do I recover this? The only way I can think for this to happen is if your source tree was inconsistent (i.e. not completely updated), or you updated your sources after you did the installworld, and the kernel depends on the newer version of config than the one you have built (in this case the 'safe' buildworld/buildkernel/installkernel method you used to begin with should still work). Kris Well, I tried makeworld again. Dies in Step 3. Reloaded all source from the distribution CD. makeworld dies in exactly the same place: Really, this all points to something else having changed on your system in the meantime. Try cd /usr/src make cleandir make cleandir make buildworld Kris Wish I had seen that earlier. I just tried a cvs from RELENG_5_3. I thought that would only have security fixes. Its downloading a ton of stuff. Obviously I can't stop it. Seems like just about all the userland source files are being changed. Lots of deletes too. I'll give the above a try when this finishes. Why so many files from cvs? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: 5.3 Building Kernel/World
On Jan 17, 2005, at 11:07, Kris Kennaway wrote: On Sun, Jan 16, 2005 at 11:32:31PM -0800, Doug Hardie wrote: The only way I can think for this to happen is if your source tree was inconsistent (i.e. not completely updated), or you updated your sources after you did the installworld, and the kernel depends on the newer version of config than the one you have built (in this case the 'safe' buildworld/buildkernel/installkernel method you used to begin with should still work). Kris Well, I tried makeworld again. Dies in Step 3. Reloaded all source from the distribution CD. makeworld dies in exactly the same place: Really, this all points to something else having changed on your system in the meantime. Try cd /usr/src make cleandir make cleandir make buildworld Kris Wish I had seen that earlier. I just tried a cvs from RELENG_5_3. I thought that would only have security fixes. Its downloading a ton of stuff. Obviously I can't stop it. Seems like just about all the userland source files are being changed. Lots of deletes too. I'll give the above a try when this finishes. Why so many files from cvs? The number of changes between RELENG_5_3_0_RELEASE and RELENG_5_3 is very small. If you're seeing lots of changes, it means that you didn't actually have a 5.3-RELEASE source tree installed before now, which explains the problems you were seeing in compiling it. Kris Thats interesting. I was using the 5.3 release CD. The checksums match those listed. It was installed onto a re-formatted drive as I wanted the UFS-2. The source was installed as part of the original installation. Anyway, either the cvsup or the cleandirs worked. I was able to buildworld and a new kernel. Installation of both appears to have gone correctly. uname gives the new kernel and strings of /boot/kernel/kernel also shows the new name. uname before said 5.3-RELEASE. It would appear that when I build the production systems I will immediately after instalation cvsup to RELENG_5_3 and then run make cleandir before anything else. Fortunately right now I am playing with test systems. It it at all possible to not have to buildworld when building a new kernel? For example, I was trying to add option atapicam. It would seem that buildworld would not be necessary in that situation. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
su from root
I have encountered an unusual issue where the behavior is different between FreeBSD 4.6 and 5.3. If I login and then su to root successfully, then do a su to a non-root user I get: pam_login_access: pam_sm_acct_mgmt: user-id is not allowed to log in on /dev/ttyv0 In chasing this down it appears that the restriction is coming from login.access which does have a limitation to prevent the non-root user from logging in. Only members of the wheel group are permitted to login. That restriction is essential to this system. However, I don't understand why su is concerned about that. I need su to switch me to that user. I suspect this may be controlled by PAM but haven't been able to figure out just where that would be. How can I make su work like it does in 4.6? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Disk Error
I have been getting the following disk errors consistently for the last month. ad2s1e: hard error reading fsbn 6934399 of 3467168-3467295 (ad2s1 bn 6934399; cn 431 tn 164 sn 52) status=59 error=40 spec_getpages:(#ad/0x20014) I/O read failure: (error=5) bp 0xc5678f94 vp 0xcb5f3a80 size: 65536, resid: 65536, a_count: 65536, valid: 0x0 nread: 0, reqpage: 0, pindex: 504, pcount: 16 vm_fault: pager read error, pid 35441 (expireover) How do you figure out which file has the problem? expireover's logs are all buffered so you don't get the last partial buffer. I don't know yet if I can mark that particular sector as bad, but if I can find the file I can at least move to someplace where it won't get deleted. I chased through the core dump and the only directory indicated but all of those files are good. I have also tar'd the entire news directory elsewhere and no errors were encountered. The sector is the same every day. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk Error
I doubt that its dying. There is only one bad sector. The drive is in constant use. Its ran at 100% for almost 12 hours while copying the files and no errors were detected. Its always the same sector with the error. On Mar 7, 2005, at 09:54, Aftab Jahan Subedar wrote: ASAP 1. fsck -y 2. tunefs ( enable softupdate) 3. backup to new hard disk 4. remove this faulty hard disk Your hard disk is dyeing . Doug Hardie wrote: I have been getting the following disk errors consistently for the last month. ad2s1e: hard error reading fsbn 6934399 of 3467168-3467295 (ad2s1 bn 6934399; cn 431 tn 164 sn 52) status=59 error=40 spec_getpages:(#ad/0x20014) I/O read failure: (error=5) bp 0xc5678f94 vp 0xcb5f3a80 size: 65536, resid: 65536, a_count: 65536, valid: 0x0 nread: 0, reqpage: 0, pindex: 504, pcount: 16 vm_fault: pager read error, pid 35441 (expireover) How do you figure out which file has the problem? expireover's logs are all buffered so you don't get the last partial buffer. I don't know yet if I can mark that particular sector as bad, but if I can find the file I can at least move to someplace where it won't get deleted. I chased through the core dump and the only directory indicated but all of those files are good. I have also tar'd the entire news directory elsewhere and no errors were encountered. The sector is the same every day. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: tech question
On Mar 6, 2005, at 23:45, Chad Leigh -- Shire.Net LLC wrote: On Mar 7, 2005, at 12:31 AM, Michael C. Shultz wrote: On Sunday 06 March 2005 11:28 pm, popbox wrote: Excuse me for foolish question and "pig latin". I'm a new user of FreeBSD and I have a trouble with mounting DVD. There is no separated information in your documentation (Handbook) about this question. I tried to mount DVD the same way as CD. It is not enough, I think. You looked at http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/creating- dvds.html This does not seem to answer the OP question. That pages deals with creating various sorts of writable DVDs. Chad I have mounted DVD-Rs numerous times on 4.6 using mount -t cd9660 /dev/ /mnt and that has worked fine. That also works on 5.3. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: how to deal with spam for good?
On Mar 10, 2005, at 01:49, Ted Mittelstaedt wrote: The only long term solution that is going to work is modding the DNS records to designate an official SMTP server for each domain, such a plan has been in the works for a while among the standard bodies that know what they are doing. SPF is only going to address one form of spam distribution. Unfortunately it does nothing for the spammers who get their own domain and establish their own SPF records. They can continue to spam away at will. Likewise SPF will not close any of the open relays run by the organizations that are pushing SPF. Those will continue to forward spam like they do today. I suspect the open relays are ahead of their SPF checking as we continue to receive mail through them even theough they claim SPF is in use. Spam will only go away when people no longer respond to it. When there is no revenue generated to cover the cost of spamming then it will end. Since spamming is so cheap, it only takes a couple of responses to cover the costs. Probability of finding a couple of morons out there is 1.00. People still respond to the Nigerian scams. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: how to deal with spam for good?
On Mar 10, 2005, at 15:24, Anthony Atkielski wrote: As it is, sometimes I can't answer clients by e-mail because their own ISPs (e.g., anything run by Time-Warner) simply throw away my e-mail because it doesn't come from a Big ISP. I doub't thats the reason. I am presuming you are referring to wanado.fr. I know we have its MTA blocked because of the unresolved spam complaints over the years. I suspect thats the same for others also. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: how to deal with spam for good?
On Mar 10, 2005, at 17:38, Anthony Atkielski wrote: Doug Hardie writes: I doub't thats the reason. I am presuming you are referring to wanado.fr. No, I'm referring to e-mail sent directly from my own server (not relayed through Wanadoo). Time-Warner and a few other ISPs either reject it openly or silently throw it away. Can't say then. However we are a fairly small ISP and Time-Warner takes our mail. I doub't size is the issue. I know we have its MTA blocked because of the unresolved spam complaints over the years. I suspect thats the same for others also. What about the millions of legitimate subscribers using this ISP? We don't receive much legitimate mail from them. Get a lot more spam. -- Anthony ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: how to deal with spam for good?
On Mar 10, 2005, at 18:30, Warren Block wrote: milter-greylist works great with sendmail. Here's a somewhat-dated article I wrote about using it and clamav-milter with sendmail: http://www.wonkity.com/greylist.pdf I am getting a no such file back on that. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Which Way to Partition.
On Jan 21, 2005, at 19:32, Greg 'groggy' Lehey wrote: On Friday, 21 January 2005 at 22:01:14 -0500, Peterhin wrote: I am new to FreeBSD, and have only used Linux for less than a year. I have read the "Handbook", also "FreeBSD An open-source system for your personal computer", they both suggest that I do a standard installation, whereas in "The Complete FreeBSD" by Greg Lehey, his suggestion is to do the custom installation. Any suggestions as to which way to go.? I recommend the custom installation. I also say why. Well, I am looking at the 3rd Edition page 71 where it appears you recommend the custom and the novice installations. The only real comment about the custom installation is that it takes you back to the top menu after each step. I have installed may copies of versions 2,3,4, and not 5 and don't see what the advantage of that might be. The only reason that comes to mind is if you botch something you can go back and redo it. That doesn't seem like much of a big deal to me, but... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Which Way to Partition.
On Jan 21, 2005, at 23:20, Greg 'groggy' Lehey wrote: [Format recovered--see http://www.lemis.com/email/email-format.html] Long/short syndrome. On Friday, 21 January 2005 at 20:58:35 -0800, Doug Hardie wrote: On Jan 21, 2005, at 19:32, Greg 'groggy' Lehey wrote: On Friday, 21 January 2005 at 22:01:14 -0500, Peterhin wrote: I am new to FreeBSD, and have only used Linux for less than a year. I have read the "Handbook", also "FreeBSD An open-source system for your personal computer", they both suggest that I do a standard installation, whereas in "The Complete FreeBSD" by Greg Lehey, his suggestion is to do the custom installation. Any suggestions as to which way to go.? I recommend the custom installation. I also say why. Well, I am looking at the 3rd Edition page 71 where it appears you recommend the custom and the novice installations. The only real comment about the custom installation is that it takes you back to the top menu after each step. I have installed may copies of versions 2,3,4, and not 5 and don't see what the advantage of that might be. The only reason that comes to mind is if you botch something you can go back and redo it. That doesn't seem like much of a big deal to me, but... It's not a big deal, but it helps. You're less likely to need to go back when you're proficient, but it doesn't harm to have the facility. It doesn't cost you anything. That makes sense. Glad to know there isn't something I missed. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Possible SCSI address conflicts
FreeBSD 5.3-P5 with device atapicam in the kernel. From dmesg.boot: Waiting 15 seconds for SCSI devices to settle da1 at ahc0 bus 0 target 6 lun 0 da1: Fixed Direct Access SCSI-3 device da1: 40.000MB/s transfers (20.000MHz, offset 8, 16bit), Tagged Queueing Enabled da1: 35074MB (71833096 512 byte sectors: 255H 63S/T 4471C) cd0 at ata1 bus 0 target 0 lun 0 cd0: Removable CD-ROM SCSI-0 device cd0: 33.000MB/s transfers cd0: Attempt to query device size failed: NOT READY, Medium not present - tray c losed cd1 at ata1 bus 0 target 1 lun 0 cd1: Removable CD-ROM SCSI-0 device cd1: 33.000MB/s transfers cd1: Attempt to query device size failed: NOT READY, Medium not present da0 at ahc0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: 40.000MB/s transfers (20.000MHz, offset 8, 16bit), Tagged Queueing Enabled da0: 4357MB (8925000 512 byte sectors: 255H 63S/T 555C) Mounting root from ufs:/dev/da1s1a dc0: failed to force tx and rx to idle state dc0: failed to force tx and rx to idle state dc0: failed to force tx and rx to idle state It looks like there is a scsi conflict. Both da0 and cd0 have the bus numbers. I won't be back on site till next Friday to try the drive. Is this an issue? Other than changing jumpers on the drives is there a way to resolve it if needed? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Reboot Hangs
FreeBSD 5.3-P5 with optionsBROKEN_KEYBOARD_RESET in the kernel. System was first built on hardware that required that option to be able to avoid hanging on reboot. However, now I have installed it on a newer system. It still has the option defined. And it hangs. I suspect that I don't need the option on this system. Is there a way to disable it without having to rebuild the system? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Disk Label Problem
I have a system with two SCSI disks. da1 has a complete working system on it that I need to clone onto da0. The disks are different sizes. So I went to sysinstall and used 'disk label' to create the desired structure. Thats where the problems started. If I create the first partition and set the mount point to / and the second as a swap partition and the third to mount at /usr then when writing the changes there are a number of errors generated because it can't mount to those points - they are in use. So then I tried to use 'disk label' and create the structure using /mnt and /mnt1 (which do exist). That worked fine and did the newfs. However, it created partitions d and e rather than a and d. So I went back and reestablished the structure using / and /usr to set the partitions to a and d and then went back and changed the mount points to /mnt and /mnt1 before the write. However, this generated an error that it couldn't write label. Obviously I am doing something wrong since I have don this using sysinstall and completing the system installation from CD. However, in this case the machine is a long way away and the CD drive is empty. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
SCSI Problem
I have a system that was running fine with 2 SCSI drives. Both on the same line, the last one terminated. I removed the first one leaving the one with the termination. Now when the system boots I get the strangest messages and the results are quite unusual. Here are the console messages during the boot process: Waiting 15 seconds for SCSI devices to settle ahc0: Recovery Initiated >> Dump Card State Begins < ahc0: Dumping Card State in Command phase, at SEQADDR 0x170 Card was paused ACCUM = 0x80, SINDEX = 0xac, DINDEX = 0xc0, ARG_2 = 0x4 HCNT = 0x0 SCBPTR = 0x0 SCSISIGI[0x84]:(BSYI|CDI) ERROR[0x0] SCSIBUSL[0x80] LASTPHASE[0x80]:(CDI) SCSISEQ[0x12]:(ENAUTOATNP|ENRSELI) SBLKCTL[0x2]:(SELWIDE) SCSIRATE[0x0] SEQCTL[0x10]:(FASTMODE) SEQ_FLAGS[0x0] SSTAT0[0x7]:(DMADONE|SPIORDY|SDONE) SSTAT1[0x2]:(PHASECHG) SSTAT2[0x0] SSTAT3[0x0] SIMODE0[0x0] SIMODE1[0xac]:(ENSCSIPERR|ENBUSFREE|ENSCSIRST|ENSELTIMO) SXFRCTL0[0x88]:(SPIOEN|DFON) DFCNTRL[0x4]:(DIRECTION) DFSTATUS[0x6d]:(FIFOEMP|DFTHRESH|HDONE|FIFOQWDEMP|DFCACHETH) STACK: 0x37 0x0 0x16a 0x19a SCB count = 20 Kernel NEXTQSCB = 1 Card NEXTQSCB = 19 QINFIFO entries: 19 18 9 0 7 6 17 8 15 14 5 4 3 2 Waiting Queue entries: Disconnected Queue entries: QOUTFIFO entries: Sequencer Free SCB List: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Sequencer SCB Info: 0 SCB_CONTROL[0x0] SCB_SCSIID[0x10] SCB_LUN[0x0] SCB_TAG[0x10] 1 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 2 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 3 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 4 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 5 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 6 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 7 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 8 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 9 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 10 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 11 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 12 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 13 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 14 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 15 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] Pending list: 2 SCB_CONTROL[0x0] SCB_SCSIID[0xf0]:(TWIN_CHNLB|TWIN_TID) SCB_LUN[0x0] 3 SCB_CONTROL[0x0] SCB_SCSIID[0xe0]:(TWIN_CHNLB) SCB_LUN[0x0] 4 SCB_CONTROL[0x0] SCB_SCSIID[0xd0]:(TWIN_CHNLB) SCB_LUN[0x0] 5 SCB_CONTROL[0x0] SCB_SCSIID[0xc0]:(TWIN_CHNLB) SCB_LUN[0x0] 14 SCB_CONTROL[0x0] SCB_SCSIID[0x90]:(TWIN_CHNLB) SCB_LUN[0x0] 15 SCB_CONTROL[0x0] SCB_SCSIID[0x80]:(TWIN_CHNLB) SCB_LUN[0x0] 8 SCB_CONTROL[0x0] SCB_SCSIID[0x70]:(TWIN_TID) SCB_LUN[0x0] 9 SCB_CONTROL[0x0] SCB_SCSIID[0x60] SCB_LUN[0x0] 18 SCB_CONTROL[0x0] SCB_SCSIID[0x30] SCB_LUN[0x0] 19 SCB_CONTROL[0x0] SCB_SCSIID[0x20] SCB_LUN[0x0] 16 SCB_CONTROL[0x0] SCB_SCSIID[0x10] SCB_LUN[0x0] 17 SCB_CONTROL[0x0] SCB_SCSIID[0xb0]:(TWIN_CHNLB) SCB_LUN[0x0] 6 SCB_CONTROL[0x0] SCB_SCSIID[0xa0]:(TWIN_CHNLB) SCB_LUN[0x0] 7 SCB_CONTROL[0x0] SCB_SCSIID[0x50] SCB_LUN[0x0] 0 SCB_CONTROL[0x0] SCB_SCSIID[0x40] SCB_LUN[0x0] Kernel Free SCB list: 13 12 11 10 Untagged Q(1): 16 Untagged Q(2): 19 Untagged Q(3): 18 Untagged Q(4): 0 Untagged Q(5): 7 Untagged Q(6): 9 Untagged Q(7): 8 Untagged Q(8): 15 Untagged Q(9): 14 Untagged Q(10): 6 Untagged Q(11): 17 Untagged Q(12): 5 Untagged Q(13): 4 Untagged Q(14): 3 Untagged Q(15): 2 < Dump Card State Ends >> (probe14:ahc0:0:11:0): SCB 0x11 - timed out sg[0] - Addr 0x174d41c0 : Length 32 (probe14:ahc0:0:11:0): Other SCB Timeout ahc0: Issued Channel A Bus Reset. 15 SCBs aborted ahc0: Timedout SCBs already complete. Interrupts may not be functioning. ahc0: Recovery Initiated >> Dump Card State Begins < Same as before < Dump Card State Ends >> (probe14:ahc0:0:11:0): SCB 0x5 - timed out sg[0] - Addr 0x174d4060 : Length 32 (probe14:ahc0:0:11:0): Other SCB Timeout ahc0: Issued Channel A Bus Reset. 15 SCBs aborted ahc0: Timedout SCBs already complete. Interrupts may not be functioning. cd0 at
Re: SCSI Problem
On Feb 5, 2005, at 15:59, Doug Hardie wrote: I have a system that was running fine with 2 SCSI drives. Both on the same line, the last one terminated. I removed the first one leaving the one with the termination. Now when the system boots I get the strangest messages and the results are quite unusual. Here are the console messages during the boot process: . From here on out the system completes booting as normal and runs just fine. Everything works properly except that the system thinks it has 16 SCSI drives. There is only one, but camcontrol shows it on all targets and disklabel gives the real disk label for all values of /dev/da0s1 through /dev/da14/s1. The physical disk has no jumpers. Any ideas what might cause this? I have never seen anything like it before. I can't imagine what I did to cause this. Here is the camcontrol devlist -v output: scbus0 on ahc0 bus 0: at scbus0 target 1 lun 0 (pass0,da0) at scbus0 target 2 lun 0 (pass1,da1) at scbus0 target 3 lun 0 (pass2,da2) at scbus0 target 4 lun 0 (pass3,da3) at scbus0 target 5 lun 0 (pass4,da4) at scbus0 target 6 lun 0 (pass5,da5) at scbus0 target 7 lun 0 (pass6,da6) at scbus0 target 8 lun 0 (pass7,da7) at scbus0 target 9 lun 0 (pass8,da8) at scbus0 target 10 lun 0 (pass9,da9) at scbus0 target 11 lun 0 (pass10,da10) at scbus0 target 12 lun 0 (pass11,da11) at scbus0 target 13 lun 0 (pass12,da12) at scbus0 target 14 lun 0 (pass13,da13) at scbus0 target 15 lun 0 (pass14,da14) < > at scbus0 target -1 lun -1 () I have made some progress. Pulling the SCSI cable and reseating the controller eliminated the error messages. However, the above devlist still occurs. The controller is an Adaptec 2940UW. The adaptec configuration software shows one disk on ID 0 and the controller on ID 7. The above listing doesn't find the disk on target 0. My other systems with the same setup do. I won't be back on site till Friday so I am looking for ideas on what to check or try. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: SSH terminal locking up from OS X to FreeBSD
On Feb 22, 2005, at 13:50, [EMAIL PROTECTED] wrote: * Eric F Crist <[EMAIL PROTECTED]> [2005-02-22 15:35:53 -0600]: On Feb 22, 2005, at 3:11 PM, [EMAIL PROTECTED] wrote: What version of Mac OS X are you using? All of my workstations are Mac OS X, and all but one server (an old cobalt raq 2) are running FreeBSD 5.3, and I have never seen a problem with using ssh from a terminal to a FreeBSD system. OS X is always the latest, currently 10.3.8. I have no control over the version this particular FreeBSD system, but this problem has persisted for several versions of Mac OS X and FreeBSD. I have been using ssh from my Macs to FreeBSD versions from 2.5 and up. Currently I have servers running 4.6 and 5.3. My connections stay active for one hour without problems. The one hour limit is from an undocumented feature in Apple's Airport that terminates a connection if there is no activity for an hour. The configuration on both ends is out of the box except that I force version 2 on both machines and have changed the port away from 22. To see what is happening using tcpdump do the following on both machines as root: 1. tcpdump -xXs1500 port 22 > xxx (xxx is some file name to save the trace) 2.open the connection and cause it to fail 3. terminate the tcpdumps. 4. The traces in the xxx files will be time stamped and you should be able to check them side by side and watch what happens. One or the other will probably stop responding. Another thing that may help is to use (as root) ktrace on the sshd server and on the ssh client. That will generate a lot of output but may help with the tcpdump to see why the problem is occuring. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: SSH terminal locking up from OS X to FreeBSD
On Feb 22, 2005, at 22:57, Jim Freeze wrote: * Chad Leigh -- Shire.Net LLC <[EMAIL PROTECTED]> [2005-02-22 22:58:17 -0700]: Just for giggles, what happens when you try a different encryption method with the ssl client? For example, -c blowfish Ok, so I tried this, but it still locks up. However, I was able to do ~C to get a command line and ~^Z to background the ssh terminal, but I was never able to re-activate it. I did manage to log the IP activity through tcp dump, and I discovered that after the 'lock up', there are no IP messages originating from the remote machine. Also, the IP blocks are of type FP, whatever that is. (Hmm, maybe I need to clear out the known hosts on the remote machine.) An abbreviated version is below. The full log file is at: http://www.freeze.org/tcpdump3b.log 00:22:59.999439 IP localhost.53245 > remotemachine.com.ssh: S 611378943:611378943(0) win 65535 00:23:00.053942 IP remotemachine.com.ssh > localhost.53245: S 77400915:77400915(0) ack 611378944 win 57344 00:23:00.054039 IP localhost.53245 > remotemachine.com.ssh: . ack 1 win 65535 00:23:00.331844 IP remotemachine.com.ssh > localhost.53245: P 1:24(23) ack 1 win 57964 00:23:04.922358 IP localhost.53245 > remotemachine.com.ssh: . ack 3512 win 65535 # Long break - remote terminal stops responding but data is still flowing as you can see. # 00:34:05.662885 IP localhost.53245 > remotemachine.com.ssh: P 1519:1559(40) ack 3512 win 65535 00:34:07.284836 IP localhost.53245 > remotemachine.com.ssh: P 1519:1559(40) ack 3512 win 65535 00:34:09.285235 IP localhost.53245 > remotemachine.com.ssh: P 1519:1559(40) ack 3512 win 65535 00:34:43.290382 IP localhost.53240 > remotemachine.com.ssh: FP 0:48(48) ack 1 win 65535 # ~? 00:35:09.294870 IP localhost.53245 > remotemachine.com.ssh: P 1519:1719(200) ack 3512 win 65535 00:37:17.308387 IP localhost.53245 > remotemachine.com.ssh: FP 1519:2655(1136) ack 3512 win 65535 #Closed terminal The localhost is trying to send the 40 bytes in its buffer. It is not receiving and ACK from remotemachine so it retries until it eventually gives up. The F flag is localhost issuing a FIN to remotemachine to drop the TCP connection. It tries a couple times and then likewise gives up. I would recommend a ktrace on the server to see if it yields any additional information. My guess is that the sshd process has died. syslog might not be set to catch the error it may be generating. ktrace will show all the syslog calls. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Port Problem
I seem to have done something to a port that is causing a problem. The port is dspam and I first did a make on it. Up cam this nice configuration option window (similar to sysinstall) where I select what turns out to be incompatable options. However, that wasn't obvious at the time. The patching and configuration completed successfully. All the various required ports installed properly. However, the make of dspam failed because of the incompatable options. The error message made it all obvious. However, I can't find a way to go back to that configuration option window to correct the problem. Make just takes me back to the compile error. Removing the work directory and the tar file results in a new download and then a silent return to the same problem. Make clean does essentially the same thing. The configuration options are being stored somewhere and I suspect I need to delete them, but where? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
I need to resend messages from dead.letters
There was a problem last night with my mail server and a bunch of mail went into the dead.letters mailbox rather than being sent. I have that mailbox and need a way to send all of those messages. I split them out into individual files, but there are just too many to send by hand. Is there a way to cause them all to be resent? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
INN Problems
I have sent the request below to the INN maillist but got no response. I have gotten nowhere trying to figure this out. Any help will be appreciated. I am running inn 2.4.0 and a few days ago postings by my users no longer get sent back to the news feed server. I have verified with them they are not receiving them from us. The postings are in the files here and can be seen by our users. Nothing apears in the outgoing file for the feed site. nntpsend.log shows the connections to the feed site, but nothing is ever sent. Traces of nnrpd and innd so no attempts to access the outgoing file. errlog, news.crit, and news.err are all empty. How can I find out what has gone wrong? Thanks, -- Doug ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: I need to resend messages from dead.letters
On Jan 13, 2004, at 04:18, Matthew Seaman wrote: On Tue, Jan 13, 2004 at 09:24:21AM +, Jez Hancock wrote: On Mon, Jan 12, 2004 at 07:23:20PM -0800, Doug Hardie wrote: There was a problem last night with my mail server and a bunch of mail went into the dead.letters mailbox rather than being sent. I have that mailbox and need a way to send all of those messages. I split them out into individual files, but there are just too many to send by hand. Is there a way to cause them all to be resent? Or to split up the dead.letter mailbox into individual numbered messages: % formail -s /bin/sh -c 'cat > msg.$FILENO' < dead.letter and you can pipe each message into sendmail as above to re-send it: % /usr/sbin/sendmail -v -t -oiee < msg.999 Nb. be careful when doing this sort of thing, or you'll spray e-mails all over the place and make yourself quite unpopular. Thanks. I had missed the -t option to sendmail. That does exactly what I needed. -- Doug ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
File deletion problem
I have a situation that I have not been able to track down where on one of my servers some process is writing a log file (I presume) and it is getting rotated out from under it. The net result is that the log continues to be written to the original file which eventually is deleted thus leaving no trace of who or what. It takes several months before its size becomes noticable, but eventually get grows to consume remaining disk space. Given that the file has an inode but no directory entry, how do you find it? All I have been able to come up with is to use fstat to find all the open files inodes and then to search with ls for each by hand and removing those I can find. Unfortunately this is a large web server with lots of files. Today I moved some of the log files onto a different disk to see if the problem moves. That would narrow down the search considerably. But I suspect I will have to wait a couple months before I can see the effects of the hidden file. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Bind 8 vs. Bind 9
I have a large mail server with a couple of zones defined where the sum of the zone definition files is 153 MB. When I use Bind 8 the VSIZE for bind jumps to 250 MB. Thats with nothing going on using bind. When I switch to Bind 9 and load the same files the VSIZE jumps to 353 MB. I was hoping to use the max-cache-size feature in bind 9 but the extra size of it makes it impractical. Why is it that much larger? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Reboot Problem with 5.2.1
On Sun, 4 Apr 2004 23:23:06 -0700 Doug Hardie <[EMAIL PROTECTED]> wrote: I am testing 5.2.1 in preperation for moving production servers eventually from 4.6 to 5.x. Most of the issues I have figured out, but there is one that I cannot get to work - shutdown -r now. Rebooting dies consistently. With the GENERIC kernel I get the message: Rebooting... Keyboard reset did not work, attempting CPU shutdown In NOTES is a dexcription of BROKEN_KEYBOARD_RESET so I added that and rebuilt the kernel. Now all I get is the Rebooting... line and nothing more. Granted the system I am using for testing is not at all like the production hardware, but rebooting worked fine on 4.6 with this system. I am very reluctant to convert any production systems unless I can be sure they can successfully be rebooted without having a person on-site. These machines are all unattended and quite far away. Is there a workaround for this issue? Try toggeling hw.acpi.disable_on_poweroff with sysctl. Regards, Stephen Hilton Setting both the BROKEN_KEYBOARD_RESET and hw.acpi.disable_on_powerff to 0 fixed my problem. Now the systems reboot properly. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Unusual login requirement
I am trying to figure out how to implement an unusual login requirement and haven't found a good approach yet. What I need is to have a specific user id that when it is logged in it executes a specific script and then immediately logs out. Basically what it needs to do is run a make that builds a CD from a bunch of files and then burns the CD. Obviously a blank CD would need to be in the burner first. I don't want a general login as this would be used by a person who should not have access to the system. I just need him to be able to burn a CD frequently. My first throught was to create a script and set it as the shell in the passwd file and add it to /etc/shells. Is that the best approach? I am not concerned about the user breaking out of the script as he is trusted. I just don't want to create a regular user account for him. The server is running FreeBSD 4.6. Thanks, -- Doug ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Changing SendMail Port Number
On Jun 5, 2004, at 16:49, Gerard Seibert wrote: This is probably a stupid question, but how do I change the SMTP port number that SendMail listens in on? I want to change it to something else, like perhaps 24. My ISP is blocking 25 and I want to get around that problem. I use the following in the mc file: DAEMON_OPTIONS(`Port=25, Name=MTA')dnl DAEMON_OPTIONS(`Port=26, Name=MTA')dnl It responds to both ports 25 and 26. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Changing SendMail Port Number
On Jun 5, 2004, at 17:35, Mark wrote: Gerard Seibert wrote: This is probably a stupid question, but how do I change the SMTP port number that SendMail listens in on? I want to change it to something else, like perhaps 24. My ISP is blocking 25 and I want to get around that problem. And how will clients (the world) find you then, on port 24? Besides, are you sure your ISP blocks *incoming* port 25? That is somewhat unusual; *outgoing* 25, yes (for dialup users), but incoming? Regardless, same difference: you can start sending on port 24, but since the world is listening on port 25, that will do you little good. There are ISPs out there that block port 25 to any destination other than their mail server. If you are connected to one of them there is no way to access your ISP's mail server. Thats why we provide support for both ports 25 and 26. I have never seen port 26 blocked. Almost all mail clients provide the ability to change the port it uses. We provide instructions to our users on how to make that change if they need it sl that they can send mail through our server. We do require the use of SMTP-AUTH to avoid an open relay. Blocking port 25 is an attempt to prevent the use of open relays. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Top Consistency
I am running FreeBSD 4.6 and top does not show consistent data (at least in my understanding). The cpu states line shows the percent of time in user state. I would expect the percent processor used by all the active processes to add up to something close to that. (single processor machine). However, it never seems to come close. Often it will show 25% user and the sums of the active processes utilizations will be around 2%. Other times it will show 2% user and the sum of the processes is over 10%. Is top wacky or is my understanding wrong? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
jabberd
Has anyone succeeded in making mu-conference work with jabberd v2 on FreeBSD 5.x? I can get jabberd working fine but it never seems to route anything to mu-conference. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: mac osx disklabels
On Jun 20, 2005, at 07:59, Chad Leigh -- Shire.Net LLC wrote: On Jun 20, 2005, at 8:12 AM, Bob Bomar wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [EMAIL PROTECTED] wrote: | I went to mount a UFS filesystem on an OSX prepared drive and discovered | that apparantly FreeBSD can't read mac disklabels? Is this true or am I | missing something? | OS X Uses HFS+ which FreeBSD can not read. Its an Apple format. There were some tools in ports to read HFS fs's, but not HFS+. OS X also supports a form of UFS btw If the drive was formatted using Disk Utility there is a very hidden option for "Apple Partitioning Scheme" or "PC Partitioning Scheme". The Apple Partitioning Scheme is the default. The only documentation I could find on those options is a note that if you want to be able to mount the drive on a PC you must use the PC Partitioning Scheme. I suspect that the Apple Partitioning Scheme uses a different format for the partition map which may not be handled by anything else. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: OFF-TOPIC but ... you will laugh !!
Not surprising. Gates and Microsoft didn't develop DOS. They bought it. On Nov 2, 2005, at 20:27, Moffatt, Chris wrote: It is a reserved word from the DOS days (like "prn") I think it stands for "console" Actually, you can't create a folder named: CON, PRN, AUX, CLOCK$, NUL, COM1, COM2, COM3, COM4, COM5, COM6, COM7, COM8, COM9, LPT1, LPT2, LPT3, LPT4, LPT5, LPT6, LPT7, LPT8, and LPT9 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Aggelos Sent: Wednesday, November 02, 2005 8:31 PM To: freebsd-questions@freebsd.org Subject: OFF-TOPIC but ... you will laugh !! An Indian discovered that nobody can create a FOLDER anywhere named as "con". This is something pretty cool...and unbelievable... At Microsoft the whole Team, including Bill Gates, couldn't answer why this happened! Try it out yourself... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions- [EMAIL PROTECTED]" ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions- [EMAIL PROTECTED]" !DSPAM:43699b10336331518010033! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Upgrade from 5.3 to 6.0
I am in the midst of upgrading via source from 5.3 to 6.0. All is going fine, but the instructions in UPDATING do not include a make installkernel command. I know that needs to be done somewhere. I suspect between the buildkernel and the reboot. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Upgrade from 5.3 to 6.0
On Nov 6, 2005, at 22:15, Hans Nieser wrote: Doug Hardie wrote: I am in the midst of upgrading via source from 5.3 to 6.0. All is going fine, but the instructions in UPDATING do not include a make installkernel command. I know that needs to be done somewhere. I suspect between the buildkernel and the reboot. I think it says "make kernel ...", which apparently does both I see that now. But, then how do you build multiple kernels? I maintain all source on one system and build all the kernels there. I don't want to install them as they won't work. Also, I don't want to build them on the production machines, just install them. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Upgrade from 5.3 to 6.0
On Nov 7, 2005, at 00:10, Hans Nieser wrote: Doug Hardie wrote: On Nov 6, 2005, at 22:15, Hans Nieser wrote: Doug Hardie wrote: I am in the midst of upgrading via source from 5.3 to 6.0. All is going fine, but the instructions in UPDATING do not include a make installkernel command. I know that needs to be done somewhere. I suspect between the buildkernel and the reboot. I think it says "make kernel ...", which apparently does both I see that now. But, then how do you build multiple kernels? I maintain all source on one system and build all the kernels there. I don't want to install them as they won't work. Also, I don't want to build them on the production machines, just install them. I think you can still use the buildkernel and installkernel targets for that purpose, they are still mentioned in the Makefile at least. I verified that is correct. Thanks. I was able to build multiple kernels successfully without having to install them all. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Backup Mail Server Questions
On Sep 27, 2004, at 11:39, Nico Meijer wrote: Regular folks don't understand how mail works. They have no clue whatsoever. They don't _want_ to have a clue either. They are just behaving like consumers, again. Do you *really* want to know what's on your plate at dinner? ;-) I do, maybe you too, but most people don't. If I had a dime for every time I have had to discuss how mail delivery actually works to Joe Average or his Windows NT/2000 systems administrator... boy. Again, I have many _very_ strong opinions on how email should be managed, this is one of them. I happen to have a very strong opinion on the grim state of humanity in general and regular, everyday, Joe Average computer users in particular. I am therefore strongly biased. ;-) When Joe Average computer user sends an order to Jane Trader to sell his stock in xxx because its the highest its ever been and that email sits in your secondary MX until after xxx falls to penny stock status, then Joe Average computer user will have plenty of world class lawyers on his doorstep with big dollar signs in their eyes. They will have no problem convincing Joe Sub-Average juror (of which there will be more than enough to go around) that you were the cause of Joe Average computer users' loss of his entire retirement savings. After all, you accepted the email and acknowledged it and failed to deliver it to Jane in a timely fashion. Any technical arguments you make about the server down etc., will not faze the judge (who couldn't care less - he gets paid the same no matter who wins) or Joe Sub-Average juror who is only interested in who is putting on the better entertainment (you or the soap opera he is missing at home). ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: IPv6 && getaddrinfo(3C)
On 12 July 2012, at 07:24, Matthias Apitz wrote:
>
> Hello,
>
> I'm playing around with IPv6 code on a FreeBSD 9 system and can't get
> getaddrinfo(3C) to do what it should do as stated in its man page:
> accept an IPv6 and IPv4 IP addr, it only works with the IPv6 form:
>
> $ ./a.out ::1
> host: ::1
> read: SSH-2.0-OpenSSH_5.6p1 FreeBSD-2010
> $ ./a.out 127.0.0.1
> host: 127.0.0.1
> ssh: getaddrinfo failed code 8: hostname nor servname provided, or not known
> $ telnet 127.0.0.1 22
> Trying 127.0.0.1...
> Connected to localhost.
> Escape character is '^]'.
> SSH-2.0-OpenSSH_5.6p1 FreeBSD-2010
>
> the used C-code is attached below; what I'm doing wrong in the code?
>
> Thanks
>
> matthias
>
> /* IPv6 client code using getaddrinfo */
>
> #include
> #include
> #include
> #include
> #include
> #include
> #include
>
>
> main(argc, argv) /* client side */
> int argc;
> char *argv[];
> {
>
> struct addrinfo req, *ans;
> int code, s, n;
> char buf[1024];
>
> memset(&req, 0, sizeof(req));
> req.ai_flags = AI_ADDRCONFIG|AI_NUMERICHOST;
> req.ai_family = AF_INET6; /* Same as AF_INET6. */
> req.ai_socktype = SOCK_STREAM;
>
> /* */
> /* Use default protocol (in this case tcp) */
> /* */
>
> req.ai_protocol = 0;
>
> printf("host: %s\n", argv[1]);
> if ((code = getaddrinfo(argv[1], "ssh", &req, &ans)) != 0) {
> fprintf(stderr, "ssh: getaddrinfo failed code %d: %s\n", code,
> gai_strerror(code));
> exit(1);
> }
>
>
> /* */
> /* ans must contain at least one addrinfo, use */
> /* the first. */
> /* */
>
> s = socket(ans->ai_family, ans->ai_socktype, ans->ai_protocol);
> if (s < 0) {
> perror("ssh: socket");
> exit(3);
> }
>
> /* Connect does the bind for us */
>
> if (connect(s, ans->ai_addr, ans->ai_addrlen) < 0) {
> perror("ssh: connect");
> exit(5);
> }
>
> n = read(s, buf, 1024);
> printf ("read: %s", buf);
>
> /* */
> /* Free answers after use */
> /* */
> freeaddrinfo(ans);
>
> exit(0);
> }
>
>
I won't claim to be an expert on this, but I have used getaddrinfo successfully
in servers. The only thing I see that might be an issue is the use of zero for
ai_protocol. The comment in the man page implies that value is for servers and
not clients. I suspect you have to set the specific protocol you want. You
haven't included AI_PASSIVE so I suspect its expecting you to use the address
to contact a server.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: how to speed up port make??
On 25 July 2012, at 23:04, Ryan Noll wrote: > Hello, > > On Jul 25, 2012 7:34 PM, "Chad Perrin" wrote: >> You kids have got it easy. I used to have to compile by hand with a pair >> of tweezers, bar copper wire, a magnifying glass, and a potato with two >> pieces of metal stuck in it as a power source. > > Ha-ha... Ah those were the days..., but does anyone remember the "old" way > of building the kernel in the 2.2.8 days? I was just getting started doing > the basic system setup/admin things in those days. Back then (1998 or so) I > did not have access to broadband, so I did not even update the sources back > then, but I knew that it was a good idea to remove devices from the GENERIC > kernel that I did not have--thanks to the book by Greg Lehey. (Even though > the version of "The Complete FreeBSD" I bought is so out of date I cannot > bring myself to throw it away--it was my guide back in those days.) > > Does anyone else remember "The Complete FreeBSD"? Its sitting in my bookshelf. Its pretty worn out though. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
freebsd-update
I am using freebsd-update to update a system running a generic kernel. I ran into an interesting situation where after it has downloaded the updates it enters a configuration phase where it shows "updated" config files with the old and new. You can hit return to enter vi and clean up the file. After that you get to a selection of files where you only get the question does this look reasonable? Your options are Y or N. Y makes the changes and N just terminates the entire update forcing you to start over again from the beginning. Why can't you correct issues with those config files? Why bother to even ask if there is only one possible response (Y)?___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
SATA Controllers
Looking through the list of SATA Controllers available at Best Buy, I don't find any of them listed on the 9.0 hardware page. I need a couple cheap ones (for non-production systems). Does anyone have recommendations? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
send-pr Submission Times
I sent a PR using send-pr earlier today. However, after having sent it and received a line that said it was submitted, I realized I didn't include my email address. Somehow I completely overlooked that. I have been waiting for it to show up in the on-line indexes, but it hasn't so far. How long does that process normally take? I am wondering if it was just dropped because of the lack of the email address. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
FreeBSD 9.1 and SU+J
I didn't notice that journaling is on by default and now dump is failing. The only way I can see to disable journaling requires that the file system be dismounted, or read-only. This is a remote machine and journaling is on root. Is there any other way that would not require me to make a long trip out to the site? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: FreeBSD 9.1 and SU+J
On 4 November 2012, at 07:04, Bas Smeelen wrote: > On 11/04/2012 03:00 PM, Bas Smeelen wrote: >> On 11/04/2012 02:11 PM, RW wrote: >>> On Sun, 04 Nov 2012 11:44:28 +0100 >>> Bas Smeelen wrote: >>> >>>> On 11/03/2012 07:30 PM, Herbert J. Skuhra wrote: >>>>> On 03.11.2012 13:48, Doug Hardie wrote: >>>>> >>>>>> I didn't notice that journaling is on by default and now dump is >>>>>> failing. The only way I can see to disable journaling requires >>>>>> that the file system be dismounted, or read-only. This is a >>>>>> remote machine and journaling is on root. Is there any other way >>>>>> that would not require me to make a long trip out to the site? > > I guess I was a little off here, it actually worked for / also > See further below for the whole story > This was all done remote with ssh > > $ mount > /dev/da0p2 on / (ufs, local, soft-updates) > devfs on /dev (devfs, local, multilabel) > /dev/da0p3 on /tmp (ufs, local, soft-updates) > /dev/da0p4 on /var (ufs, local, soft-updates) > /dev/da0p5 on /usr (ufs, local, soft-updates) > $ su > Password: > root@osebart:/usr/home/Freebee # rm /.sujournal > root@osebart:/usr/home/Freebee # rm /var/.sujournal > root@osebart:/usr/home/Freebee # rm /tmp/.sujournal > root@osebart:/usr/home/Freebee # rm /usr/.sujournal > root@osebart:/usr/home/Freebee # uname -a > FreeBSD osebart.ose.nl 9.1-RC2 FreeBSD 9.1-RC2 #0 r241106: Mon Oct 1 > 18:26:44 UTC 2012 > r...@farrell.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC amd64 I can't get that to work on i386. Here is /etc/rc.d/fsck: fi echo "Ready for tunefs" /sbin/tunefs -j disable /dev/da0p2 } load_rc_config $name run_rc_command "$1" reboot computer and here is the output from messages: Nov 4 14:07:19 Router kernel: Ready for tunefs Nov 4 14:07:19 Router kernel: Clearing journal flags from inode 4 Nov 4 14:07:19 Router kernel: tunefs: soft updates journaling cleared but soft updates still set. Nov 4 14:07:19 Router kernel: tunefs: remove .sujournal to reclaim space Nov 4 14:07:19 Router kernel: Mounting local file systems:. and the output from mount: Router# mount /dev/da0p2 on / (ufs, local, journaled soft-updates) devfs on /dev (devfs, local, multilabel) Journaled is still on after 2 reboots. Router# uname -a FreeBSD Router 9.1-RC2 FreeBSD 9.1-RC2 #0 r241133: Tue Oct 2 17:11:45 UTC 2012 r...@obrian.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC i386 -- Doug ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: WARNING: FreeBSD 9.0-RELEASE is approaching its End-of-Life date.
On 15 November 2012, at 14:46, Matthias Petermann wrote: > Hi, > > On Thu, 15 Nov 2012 14:35:52 -0800 > Michael Sierchio wrote: > >> http://www.freebsd.org/security/ >> >> Scroll down about halfway. 9.0 is a regular release, EOL is January 31, >> 2013. >> >> Alternate releases are extended releases, so 9.1 will have a 2 year >> support span. > > Thanks for the clearification. One technical thing: is it possible, to upgrade > from FreeBSD 9.0 to 9.1 with the freebsd-update utility? Yes. I have done that from 9.0 to 9.1-RC1 and later RC2. It takes longer than you would like, but works just fine. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: WARNING: FreeBSD 9.0-RELEASE is approaching its End-of-Life date.
On 15 November 2012, at 17:04, Randal L. Schwartz wrote: >> "Andreas" == Andreas Rudisch <"cyb."@gmx.net> writes: > > Andreas> On Thu, 15 Nov 2012 23:46:53 +0100 > Andreas> Matthias Petermann wrote: > >>> Thanks for the clearification. One technical thing: is it possible, to >>> upgrade >>> from FreeBSD 9.0 to 9.1 with the freebsd-update utility? > > Andreas> Yes, it is. > > Can I go from 8.3 directly to 9.1, or should I stop over at 9.0 first? For me that was not possible. My disks were partitioned and labeled when FreeBSD 4.7 was new. The size of the root partition was now too small for 9.0. I had to do a complete install and reformat of the drives to get to 9.0. My root partition was a bit small for 7.x as I had to delete the symbol files to make it fit. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: I Guess I Don't Understand NFS As Well As I Thought
On 24 November 2012, at 12:32, Tim Daneliuk wrote: > Can someone kindly explain what is going on here: > > Machine A: FreeBSD - was running 8, just upgraded to 9.1-PRE >(I don't recall seeing the behavior described below > in V8, but then, I don't think I ever tried it). > > Machine B: Linux Mint Desktop > > - Machine A acts as an NFS server for Machine B. > > - Machine A exports a particular directory like this: > > /usr/foo -maproot=myid -network ... > > > - /usr/foo/bar is owned by root on Machine A and has files therein > owned as root:root with permissions of 600. > > - If I access /usr/foo/bar/file1 from Machine B, I cannot read it > but - and this is the part I don't get - I CAN *rename* it. > > What's going on? Since /foo/bar/ is owned by root and everything > in it is 600 root:root, I would not expect a remote access to allow > things like renaming. Clearly I am missing something here, but I > don't get it. What are the permissions on the directory /usr/foo/bar? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: I Guess I Don't Understand NFS As Well As I Thought
On 24 November 2012, at 14:37, Tim Daneliuk wrote: > On 11/24/2012 03:25 PM, Doug Hardie wrote: >> >> On 24 November 2012, at 12:32, Tim Daneliuk wrote: >> >>> Can someone kindly explain what is going on here: >>> >>> Machine A: FreeBSD - was running 8, just upgraded to 9.1-PRE >>>(I don't recall seeing the behavior described below >>> in V8, but then, I don't think I ever tried it). >>> >>> Machine B: Linux Mint Desktop >>> >>> - Machine A acts as an NFS server for Machine B. >>> >>> - Machine A exports a particular directory like this: >>> >>> /usr/foo -maproot=myid -network ... >>> >>> >>> - /usr/foo/bar is owned by root on Machine A and has files therein >>> owned as root:root with permissions of 600. >>> >>> - If I access /usr/foo/bar/file1 from Machine B, I cannot read it >>> but - and this is the part I don't get - I CAN *rename* it. >>> >>> What's going on? Since /foo/bar/ is owned by root and everything >>> in it is 600 root:root, I would not expect a remote access to allow >>> things like renaming. Clearly I am missing something here, but I >>> don't get it. >> >> What are the permissions on the directory /usr/foo/bar? > > 775 > > > Let me correct something. The files in that directory are > owned by root:wheel (not root:root - I got my *nixes > confused), but they definitely have 600 perms. > > On Machine A, user 'myid' is IN the wheel group but I still > don't see how he's getting permission to rename the file.\ Renaming a file does not change the file itself. It updates the directory. Any user in group wheel has the authority to write to the directory (e.g., change a file's name). The directory permissions are rwx for group wheel. You can either try a user on machine B who is not in group wheel or change the directory permissions to 755 on /usr/foo/bar. Then it would work as you expect. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Is FreeBSD 9 Production Ready?
On 24 November 2012, at 16:36, Tim Daneliuk wrote: > On 11/24/2012 05:58 PM, Erich Dollansky wrote: >> Hi, >> >> On Sat, 24 Nov 2012 10:38:35 -0600 >> Tim Daneliuk wrote: >> >>> I am currently running FBSD 8.3-STABLE on a production server that >>> provides http, dns, smtp, and so on for a small domain. This is not >>> a high arrival rate environment but it does need to be rock solid >>> (which FBSD 4-8 have been). >> >> why would you like to break a running system? > > That's exactly what I don't want to do. > >>> >>> I am contemplating moving to the FBSD 9 family. Is this branch ready >> >> I would stay with 8.x until the end of its support and move only then >> to a new branch. It could be then 9.x or 10.y. I would then - but only >> then - prefer the 10.y branch. >> >> I retired my 7.4 only because of lightning strike this spring. >> >> Robustness is my main goal here. Any change which brings only the risk >> is avoided. > > I used to take this approach. However, I discovered the pain of fixing > a configuration that jumped several major releases was way higher than > tracking them each as they became stable. I did the 9.1-PRE upgrade today > and - once the new system was compiled and ready to be installed - had > only very minor conversion issues. > > In my case, the most painful part of conversion is the mail infrastructure. > The > server in question is the domain's mail server and it has a LOT of moving > parts with custom configurations: sendmail, greylisting, mailscanner, spam > assassin, mailman, SASL ... That is pretty much always what breaks. Doing > smaller "leaps" tends to make this more tractable to control. I am in a similar situation. Reliability is more important than anything else. I run similar mail configurations on one server, although I use different machines for incoming and outgoing mail. Jumps across versions have been more difficult. I have kept records of the steps I used for each upgrade and theose help me prepare for the next one. I am in the middle of jumping from 7.2 to 9.1. One machine is completely converted and working just fine. I had reliability problems with 9.0. It kept rebooting or crashing every few days. I am on 9.1-RC2 at the moment and its been up and working for 34 days now. I will upgrade it to 9.1 when its released. This one had to be upgraded early because it was new hardware. The old machine completely died. I have another server also running 9.1-RC2 but it is not moved into production yet. It is primarily a news server and has a large news cache that has to be moved. I am waiting for 9.1 for that. On some of my test machines I have found that 9.1 is the first release to support the built-in wireless NICs. The "service" command is really helpful. I frequently can't remember which service is in etc and which in /usr/local/etc. The largest problem I encountered in the upgrade was the disk structure. My disks were setup when using FreeBSD 3.5/3.7. As a result, the root partition is way too small today. I was able to shoe horn 7.2 in by deleting the kernel symbol files while they were being installed. 9.0/9.1 just didn't fit at all. Restructuring the disks is a time consuming job and fairly error prone in getting everything back that is needed to run production. There is also the issue that the default formatting uses SU+J which is not compatible with dump live filesystems. Now I am going to have to find the time to bring the systems down to remove journaling with no one on-site who has a clue what they are doing. I currently have 9.1-RCx running on 5 systems and have not had any stability issues with it. One system is in production but the others are lightly used. One of them is a 200 MHz machine with either 32 Meg or 64 Meg memory. It seems to be faster then when it ran 8.2 but I haven't actually done any measurements. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Problem upgrading to 9.1-Release
I have upgraded my development system to 9.1 without any problems. This system maintains kernel source and I build a new kernel with a couple extra options there. The other systems mount /usr/src and /usr/obj from it and do the install. The first one to be upgraded had no problem with make installkernel. Rebooted and ran mergemaster -p just fine. However make installworld dies within a couple seconds with the following error: install -o root -g wheel -m 444 libc_pic.a /usr/lib gencat be_BY.UTF-8.cat /usr/src/lib/libc/nls/be_BY.UTF-8.msg gencat: No such file or directory *** [be_BY.UTF-8.cat] Error code 1 /usr/bin/gencat exists. However, ktrace of the make shows: 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/usr/obj/usr/src/tmp/legacy/usr/sbin/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/usr/obj/usr/src/tmp/legacy/usr/bin/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/usr/obj/usr/src/tmp/legacy/usr/games/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/usr/obj/usr/src/tmp/usr/sbin/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/usr/obj/usr/src/tmp/usr/bin/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/usr/obj/usr/src/tmp/usr/games/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) 3347 make NAMI "/tmp/install.CuIzLuBX/gencat" 3347 make RET execve -1 errno 2 No such file or directory 3347 make CALL write(0x2,0x28c48c00,0x6) 3347 make GIO fd 2 wrote 6 bytes "gencat" Obviously its not in any of those places. How can I fix this? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
SOLVED: Problem upgrading to 9.1-Release
I found the problem. Somehow /usr/obj was not successfully exported and hence was completely empty. There must have been some error message in that process that I missed. Anyway, correcting that problem so that /usr/obj was available fixed the problem. On 4 January 2013, at 15:38, Doug Hardie wrote: > I have upgraded my development system to 9.1 without any problems. This > system maintains kernel source and I build a new kernel with a couple extra > options there. The other systems mount /usr/src and /usr/obj from it and do > the install. The first one to be upgraded had no problem with make > installkernel. Rebooted and ran mergemaster -p just fine. However make > installworld dies within a couple seconds with the following error: > > install -o root -g wheel -m 444 libc_pic.a /usr/lib > gencat be_BY.UTF-8.cat /usr/src/lib/libc/nls/be_BY.UTF-8.msg > gencat: No such file or directory > *** [be_BY.UTF-8.cat] Error code 1 > > /usr/bin/gencat exists. However, ktrace of the make shows: > > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/usr/obj/usr/src/tmp/legacy/usr/sbin/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/usr/obj/usr/src/tmp/legacy/usr/bin/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/usr/obj/usr/src/tmp/legacy/usr/games/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/usr/obj/usr/src/tmp/usr/sbin/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/usr/obj/usr/src/tmp/usr/bin/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/usr/obj/usr/src/tmp/usr/games/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL execve(0xbfbfd1c8,0x28c35f14,0x28421180) > 3347 make NAMI "/tmp/install.CuIzLuBX/gencat" > 3347 make RET execve -1 errno 2 No such file or directory > 3347 make CALL write(0x2,0x28c48c00,0x6) > 3347 make GIO fd 2 wrote 6 bytes > "gencat" > > Obviously its not in any of those places. How can I fix this? > > > ___ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org" > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Booting Problem
I have a relatively old machine that I am trying to boot 9.1 on. The bios will not boot from USB stick. I am using an external CD drive. It starts the boot process fine and gets to the Bootstrap loader message with revision 1.1. Then it puts out the machine, date, time the CD was created and starts the spinner. It spins around about 2 times and stops. The system continues to read from the drive for another couple minutes. Then everything stops. Nothing more happens. The CD is good. I can boot it just fine using the same external drive on another machine. While I could remove the drive and temporarily mount in in the working machine and build it there, I would like to find a way to successfully boot from CD. This will become a remote machine and taking it apart later is not a viable option. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Booting Problem
On 29 January 2013, at 07:18, Mario Lobo wrote: > On Mon, 28 Jan 2013 15:16:14 -0800 > Doug Hardie wrote: > >> I have a relatively old machine that I am trying to boot 9.1 on. The >> bios will not boot from USB stick. I am using an external CD drive. >> It starts the boot process fine and gets to the Bootstrap loader >> message with revision 1.1. Then it puts out the machine, date, time >> the CD was created and starts the spinner. It spins around about 2 >> times and stops. The system continues to read from the drive for >> another couple minutes. Then everything stops. Nothing more happens. >> >> The CD is good. I can boot it just fine using the same external >> drive on another machine. While I could remove the drive and >> temporarily mount in in the working machine and build it there, I >> would like to find a way to successfully boot from CD. This will >> become a remote machine and taking it apart later is not a viable >> option. ___ >> freebsd-questions@freebsd.org mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-questions To >> unsubscribe, send any mail to >> "freebsd-questions-unsubscr...@freebsd.org" > > Can you boot a different OS (Win, Ububtu, gparted, etc ...) from the > same drive on the same machine? Not so far. The drive works fine on other systems. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Booting Problem
On 29 January 2013, at 20:25, d...@safeport.com wrote: > > On Tue, 29 Jan 2013, Doug Hardie wrote: > >> On 29 January 2013, at 07:18, Mario Lobo wrote: >> >>> On Mon, 28 Jan 2013 15:16:14 -0800 >>> Doug Hardie wrote: >>> >>>> I have a relatively old machine that I am trying to boot 9.1 on. The >>>> bios will not boot from USB stick. I am using an external CD drive. >>>> It starts the boot process fine and gets to the Bootstrap loader >>>> message with revision 1.1. Then it puts out the machine, date, time >>>> the CD was created and starts the spinner. It spins around about 2 >>>> times and stops. The system continues to read from the drive for >>>> another couple minutes. Then everything stops. Nothing more happens. >>>> >>>> The CD is good. I can boot it just fine using the same external >>>> drive on another machine. While I could remove the drive and >>>> temporarily mount in in the working machine and build it there, I >>>> would like to find a way to successfully boot from CD. This will >>>> become a remote machine and taking it apart later is not a viable >>>> option. > > What is the system you are using? What external devices does it have built-in > support for? In the absence of any data - how about trying an external hard > drive? 9.1 release - Generic. Basically the disk1. Don't have an extra external drive. > > Why not remove the hard drive, use another system to put FreeBSD on the > drive, and put it back. From that point on you should be able to use the > network to upgrade. I have done that before and it does work. However, with the various changes to the system, the root partition I had previously built that way for 8.2 is just not large enough for 9.1. Also, I wanted to go to a single partition (the 9.1 default). Probably freebsd-update will take me through major releases after this, but I was hoping for a better solution so I could avoid having to transport the machine a long way twice to be able to update it. > > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Booting Problem
On 30 January 2013, at 05:16, Fbsd8 wrote: > Doug Hardie wrote: >> On 29 January 2013, at 07:18, Mario Lobo wrote: >>> On Mon, 28 Jan 2013 15:16:14 -0800 >>> Doug Hardie wrote: >>> >>>> I have a relatively old machine that I am trying to boot 9.1 on. The >>>> bios will not boot from USB stick. I am using an external CD drive. >>>> It starts the boot process fine and gets to the Bootstrap loader >>>> message with revision 1.1. Then it puts out the machine, date, time >>>> the CD was created and starts the spinner. It spins around about 2 >>>> times and stops. The system continues to read from the drive for >>>> another couple minutes. Then everything stops. Nothing more happens. >>>> >>>> The CD is good. I can boot it just fine using the same external >>>> drive on another machine. While I could remove the drive and >>>> temporarily mount in in the working machine and build it there, I >>>> would like to find a way to successfully boot from CD. This will >>>> become a remote machine and taking it apart later is not a viable >>>> option. ___ >>>> freebsd-questions@freebsd.org mailing list >>>> http://lists.freebsd.org/mailman/listinfo/freebsd-questions To >>>> unsubscribe, send any mail to >>>> "freebsd-questions-unsubscr...@freebsd.org" >>> Can you boot a different OS (Win, Ububtu, gparted, etc ...) from the >>> same drive on the same machine? >> Not so far. The drive works fine on other systems. >> > > You said in your orginal post "The bios will not boot from USB stick." > I see no reason why you would think your PC would BOOT from any USB attached > devices. > > Since you have another PC that does boot off of usb cd drive, swap hard > drives and use that pc to load FreeBSD to the hard drive. This method will > work for you. Yes that works now. But starting this weekend it will be about 100 miles away. That no longer will be practical. > > > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Unusual TCP/IP Packet Size
Monitoring a tcpdump between two systems, a FreeBSD 9.1 system has the following interface: msk0: flags=8843 metric 0 mtu 1500 options=c011b ether 00:11:2f:2a:c7:03 inet 10.0.1.199 netmask 0xff00 broadcast 10.0.1.255 inet6 fe80::211:2fff:fe2a:c703%msk0 prefixlen 64 scopeid 0x1 nd6 options=29 media: Ethernet autoselect (100baseTX ) status: active It sent the following packet: (data content abbreviated) 02:14:42.081617 IP 10.0.1.199.443 > 10.0.1.2.61258: Flags [P.], seq 930:4876, ack 846, win 1040, options [nop,nop,TS val 401838072 ecr 920110183], length 3946 0x: 4500 0f9e ea89 4000 4006 2a08 0a00 01c7 E.@.@.*. 0x0010: 0a00 0102 01bb ef4a ece1 680b ae37 1bbc ...J..h..7.. 0x0020: 8018 0410 3407 0101 080a 17f3 8ff8 4...……. The indicated packet length is 3946 and the load of data shown is that size. The MTU on both interfaces is 1500. The receiving system received 3 packets. There is a router and switch between them. One of them fragmented that packet. This is part of a SSL/TLS exchange and one side or the other is hanging on this and just dropping the connection. I suspect the packet size is the issue. ssldump complains about the packet too and stops monitoring. Could this possibly be related to the hardware checksums? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Water Damage
My church had a fire in the computer room today. The equipment was not directly damaged by the fire as the sprinkler system put it out very quickly. However, the sprinklers ran directly on the equipment for a couple hours. There are several servers, routers, hubs etc. Most of them had water pouring out when we picked them up. All but one spare router were on during this. I have carefully dried out all the units. However, one of the hubs appears to be toast. Some of the burning residue fell down and was pulled into the hub by the fan and is imbedded into some of its chips. I didn't bother with cleaning that one up. However, there is no visible damage to the remaining gear. I am letting it sit tonight and will try a power cycle on it tomorrow. Presuming that any of it is still working, the question is can it be trusted for unattended operations anymore? While the cost of most of it is not significant, the configuration time is. It would be much easier to use it rather than set up new gear. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Water Damage
Thanks for all the suggestions. Here is the latest update. The water from the sprinklers was purer than that from the tap. There was no residue from it anywhere. A bit of head (oven and hair drier used) and it was easily evaporated. However, all of the units except for one router were powered on and in use. The 2 hubs were directly below the fire and burning strands of something fell down and were sucked into them by their fans. The strands were hot enough that the melted into the chip bodies. I didn't hold much hope for them and was not surprised. Neither showed any form of life. Not even the fans came on. Also keep in mind that the ethernet cables came down from the ceiling and had no excess so water running down them had a straight forward path directly into the RJ-45 jacks. The operating router's sealed power brick is totally dead. Since its watertight, something obviously failed in the router and shorted out the brick. Trying another brick in that router caused every light on it to come on. It didn't do anything but light the lights. The non-operating router works fine. The one server that I have responsibility for (mailserver running FreeBSD 4.6) took awhile to get rewired properly. When it was yanked out, some of the internal cables were disconnected. Had to find the motherboard book to figure out how to set them back up properly. Once that was done, the machine came up and worked fine. However, its inlet fan was severly disfigured by the falling burning stuff. Since its at the bottom of the unit, the junk only marred the bottom of the frame. There were no electronics there for it to damage. The fan sounds funny now and I wouldn't trust it. However, the keyboard connector is now defective. You can't plug a keyboard into it. I couldn't find anything visibly wrong with it, it just doesn't work. I have no idea how that happened since there was a keyboard plugged in during the flooding. My only guess is that whoever unplugged it did so via the grab case and run method - leaving the keyboard to catch and disconnect itself. None of the MS servers survived. None had backups either. I suspect that will be a significant problem. However, I do have backups for the mail server and did recover the complete disk and dumped it to my laptop so that will be a simple restore. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Water Damage
On Tuesday, Dec 31, 2002, at 02:32 US/Pacific, Rob O'Donnell wrote: If it's a PS/2 type keyboard connector (small plug) there is a plastic pin that often gets broken off and left in the socket if connectors are pulled out violently, blocking a new keyboard being inserted. (Seen it often with mice.) If this is so, I've had success getting them out by using 'blue tack' (a semi-adhesive goo used to hold the kids drawings on the wall) on the end of a matchstick to grab hold of it. Right on. Thats exactly what happened. I guess I didn't have enough light to see that yesterday. I didn't get a chance to pop it out as the insurance adjuster arrived and is going over everything now. Thanks for the info. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
BIND configuration problem
I am trying to setup a master DNS server on a test network (not connected to the internet). The network has an address of 10.0.1.xxx as that happend to require the least setup. However, I am unable to get the reverse DNS file to load properly. The error messages are: Jan 5 14:59:27 freebie named[469]: home.net.rev:6: SOA for "10.in-addr.arpa" not at zone top "1.0.10.in-arpa.arpa" Jan 5 14:59:27 freebie named[469]: Zone "1.0.10.in-arpa.arpa" (file home.net.rev): no NS RRs found at zone top I have tried using 1.0.10.in-arpa.arpa and 10.in-arpa.arpa (example above). Obviously neither is correct. The forward DNS file loads correctly and resolves properly.Line 6 of the rev file is: 10.in-addr.arpa.IN SOA home.net. ops.lafn.ORG. ( To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Determining Ram
On Sunday, Feb 2, 2003, at 18:39 US/Pacific, Dragoncrest wrote: Cool. That worked. A little more info than I wanted to sort through, but now that I know about that, I now have more information to pick through later on should I need any of that information that Dmesg listed. At 01:02 AM 2/3/03 +, David Larkin wrote: Dragoncrest wrote: > I've got a rather odd question, but I'm looking for the easiest way to > determin how much ram I have on a given system without rebooting it. I'm > sure that there is some kind of console command that tells me that info, > but I have no idea where to begin looking to find out. Does anybody > know? Thanks. > use the command dmesg If your machine has been running too long the boot info will no longer be available through dmesg. However, it is retained in /var/run/dmesg.boot. That will always show the boot messages from the previous boot. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Airsnort
At 1511 -0400 7/25/2002, Justin L.Boss wrote: >Just worndering if someone has been able to get airsnort working >with FreeBSD using a Cisco airownet 350? > I have it working with the 340 if thats of any interest. -- -- Doug To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Secure FTPd
On Sunday, Oct 6, 2002, at 10:50 US/Pacific, Socketd wrote: > I have read about adding SSL support to ftpd, but I can't remember > where > I read it. I am running a ftp server using the ftpd in the base system > and now I want to only allow encrypted ftp connections. What should I > do? > Use /usr/ports/security/stunnel, to make universal SSL support to POP3, > IMAP and FTP? Or is there a better way? (I don't want to use ssh's > ftpd). > > Can I also use SSL with SMTP? I read that it was done once, but people > don't use it anymore? The problem with adding SSL to ftpd is the clients. You would have to create an ftp client with SSL added also. ssh's sftp has that capability and there are 2 generally available clients - sftp and scp. I believe there are clients for most computers. qpopper provides SSL for POP3 which works with most of the common mail clients. You may have to provide a popper port for both 110 and 995 in order to pick up both the older and newer clients. I have had to provide both. SSL can be used with sendmail. There is a lot of information available at www.sendmail.org. I have not tried that yet. Its on the list of things to do someday. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Secure FTPd
On Sunday, Oct 6, 2002, at 15:05 US/Pacific, Socketd wrote: >>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< > > On 10/6/02, 11:52:16 PM, Doug Hardie <[EMAIL PROTECTED]> wrote regarding > Re: > Secure FTPd: > >> The problem with adding SSL to ftpd is the clients. You would have to >> create an ftp client with SSL added also. ssh's sftp has that >> capability and there are 2 generally available clients - sftp and scp. >> I believe there are clients for most computers. > > Well, all my users use windows and there a some ftp clients that loves > SSL (like CuteFTP). You would have to emulate their SSL interface - which is that provided by ssh's sftpd. It could be done, but would take some research. > >> qpopper provides SSL for POP3 which works with most of the common mail >> clients. You may have to provide a popper port for both 110 and 995 >> in >> order to pick up both the older and newer clients. I have had to >> provide both. > > Ok. Performance-wise all services should run their own SSL support, but > it there one for the default ftpd? Not that I am aware of. Everyone appears to be using ssh. However, it is lacking chroot support. > >> SSL can be used with sendmail. There is a lot of information >> available >> at www.sendmail.org. I have not tried that yet. Its on the list of >> things to do someday. > > Ok, but it is not widely used? No idea. The clients supposedly support it but I have never tried it. > > Br > socketd > > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Secure FTPd
On Sunday, Oct 6, 2002, at 15:20 US/Pacific, Socketd wrote: >>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< > > On 10/7/02, 12:09:14 AM, Doug Hardie <[EMAIL PROTECTED]> wrote regarding > Re: > Secure FTPd: > >>> Well, all my users use windows and there a some ftp clients that >>> loves >>> SSL (like CuteFTP). > >> You would have to emulate their SSL interface - which is that provided >> by ssh's sftpd. It could be done, but would take some research. > > Eehhh? CuteFtp can use SSL, so when they want to connect, cuteftp first > handle the SSL setup and then acts like a normal ftp client. Could be. I haven't chased through ssh well enough to know how they do it. It would be handy to have a SSL ftpd so if you do it, make it available. > >>> Ok. Performance-wise all services should run their own SSL support, >>> but >>> it there one for the default ftpd? > >> Not that I am aware of. Everyone appears to be using ssh. However, >> it >> is lacking chroot support. > > Jep, sadly! > > Br > socketd > > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Secure FTPd
On Monday, Oct 7, 2002, at 17:18 US/Pacific, Eric Parusel wrote: > Hmm, I think you two *may* be doing down the wrong path... > There's a (proposed) standard for encrypted FTP, it's called > FTP over TLS ... > > Here's a link: > http://www.ford-hutchinson.com/~fh-1-pfh/ftps-ext.html > A number of proposed approaches for secure login and ftp have been floated over the years. Only scp, sftp, and sshd have made it into the FreeBSD base. I will keep watching. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: SSH/FTP Access
On Wednesday, Oct 9, 2002, at 21:28 US/Pacific, [EMAIL PROTECTED] wrote: > Just wondering is there a way to limit SSH access (when adding a user > or period) so that user can only use SSH to access or effect their > home directory? Not with the installed sshd > > Also is there a way to give (and limit) a user FTP access to another > users home directory? Yes. add the user ids or groups to /etc/ftpchroot. See the manpage. That will restrict users to their home directory and its sub directories. > > Thanks! > > To Unsubscribe: send mail to [EMAIL PROTECTED] > with "unsubscribe freebsd-questions" in the body of the message > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Max Email Users
On Tuesday, Oct 29, 2002, at 06:03 US/Pacific, Matt Delaine wrote: We are running FreeBSD 4.6 on a PIII 600 with 256 Meg RAM as our mail server. At what point (how many users) will we start running into trouble (have problems allowing us to send and receive email?) Thanks. I run an ISP and was using a PIII 200 MHz machine with 512 Meg Ram and supporting around 4000 active email accounts. It also handled outgoing mail, our admin functions, name service, YP master and some other low usage functions. I recently upgraded to a newer machine because it was available an had more disk space. With the old machine, I only say idle times under 90% when a user had their POP3 client set to not delete mail from the server and their mailbox grew to 100 MB or so. Then the POP3 server has to do a lot of I/O to get to the new messages. The issue is not so much the disk space as the time it takes to wade through all the old stuff. I try to convince users to correct their configurations. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: Using iBook OS X 10.2 CD Writer to create a FBSD on Intel Boot CDROM
On Wednesday, Oct 30, 2002, at 12:41 US/Pacific, paul beard wrote: Ev Batey WaSixCre wrote: Subj is the question .. Where can I find a map of how I build a Unix (esp F.BSD) CD Using Apple iBook running OSX 10.2 CD-R / CD-RW burner. All clues are welcome. Or how to overcome us govt politics ... /Everett/ man mkisofs to learn how to make a disk image and then burn that with whatever Apple provides. The image should be mountable with DiskCopy: that will indicate if it's what you want. Once you have a iso format from mkisofs you can use Toast to burn the CD. I use that approach often. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
vm error
I am getting a rash of vm errors that started today: vm_page_cache: attempting to cache busy page I don't seem to find anything obviously wrong in the system. How do I tell which process is causing the problem? It looks like something is hung, but I don't see any obvious candidates. Everything is working file and there are no obviously hung processes. The vm_page_cache module shows that the indicated condition is occuring, but no additional info. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
File Counts
How do I get a count of the files in directories? I need to be able to get a listing of the number of files in a directory and counts for the files in each sub-directory. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: File Counts
Thanks to all who responded. The approach below does just what I needed. On Monday, Dec 2, 2002, at 12:02 US/Pacific, Nathan Kinkade wrote: On Mon, Dec 02, 2002 at 02:42:28PM -0500, Kliment Andreev wrote: How do I get a count of the files in directories? I need to be able to get a listing of the number of files in a directory and counts for the files in each sub-directory. % ls -l | wc -l(In a directory) % ls -lR | wc -l (Including sub-directories) Or, if you are looking for subtotals, something close to this might be helpful. Beware that this will include a count for the "." and ".." entries. $ for dir in `find . -type d`; do echo $dir ; ls -l $dir | wc -l; done There is probably a better way to do this. Nathan To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message -- Doug To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Mail resending
One of our system accounts had all its mail blocked and there now are over 500 emails in dead.letter that need to be resent. Is there a way to send them (either from dead.letter or from separate files) without having to do each one individually? I haven't been able to find any way using mail or sendmail. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
encryption of data elements
I need to encrypt small data elements. These elements run from about
16 to 64 bytes in size. It would be really handy if the encrypted size
were the same as the original size. However, if it can't be I do need
to be able to predict the encrypted size in advance. Digging around
through openssl I came up with the following approach:
#include
#include
RC4_KEY key;
char buf[1000];
char out[2000];
int main (int argc, char *argv[])
{
int size, i;
size = strlen (argv[2]);
RC4_set_key (&key, size, argv[2]);
RC4 (&key, strlen(argv[1]), argv[1], out);
printf ("%s", out);
}
Where the first arg is the value and the second is the key. It seems
to work but I don't know if this is the best algorithm to use or if
there is a better approach. Thanks,
-- Doug
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-questions" in the body of the message
Re: HOWTO upgrade a FreeBSD production server
On Sunday, Feb 23, 2003, at 00:21 US/Pacific, Matthew Seaman wrote: On Sat, Feb 22, 2003 at 11:15:25PM -0800, Daxbert wrote: I'd like to find something a little quicker that doesn't require so much free space on the production server. I've looked at DESTDIR and creating a tarball of that directory on the staging server, but I've had problems during extraction with files and the 'schg' flag. About the most effective way of doing this is to NFS mount the /usr/src and /usr/obj heirarchies from your build server onto your production server in order to run the installations. This will create all of the installed system with the correct schg flags applied, although you may see some warnings because you can't apply such flags over NFS. However that shouldn't stop the installations working. I have a number of FreeBSD production servers on an ethernet connected to the internet. I also have a test system that is connected to the production servers via a second ethernet. That way my source (including ports) only resides on the test system. To avoid problems during the update itself, I first disconnect the production ethernet so only my machines can connect to the machine being updated. Then I mount /usr/src, /usr/obj, and /usr/ports from the test machine on the production machine. Do a make installkernel, make installworld and reboot on the production machine. Then you need to do mergemaster and most likely reboot again before reconnecting the production ethernet. Do run mergemaster on the test system first and keep a log of what you do with the files that require updating so the mergemaster on the production systems goes faster. I have been doing this for several years and never had a flags or permissions problem. -- Doug To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: [OT] file synchronization between two machines
On Tuesday, Mar 25, 2003, at 08:01 US/Pacific, Louis LeBlanc wrote: Hey all. Sorry for the OT question, but here goes. Anyone know of a tool or method that can check the last modification date of two files under these conditions and keep them in sync? I've never tried this, but you might give rsync with the -u option a try (test it first on unimportant files). I believe you would need to run it on both machines as it would only update in one direction. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
Re: FreeBSD upgrade maintenance vs. debian (please help)
At 0948 -0600 7/22/2002, David Wilk wrote: >How do you guys deal with this? warm-failover systems to take over during >downtimes? Or do you just accept that the system will go down for a while >at least once/year? > I run an ISP that uses FreeBSD for all our servers. I keep one additional "test/development" server that is not used for production. It has all the source/ports etc. New versions are loaded on it first. All the makeworld is done on it. The various applications are tested. Once I am convinced everything is working, then I bring down a production server, NFS mount /usr/src, /usr/obj, and /usr/ports from the test machine and do a make installworld and make installkernel. Depending on the update, mergemaster may need to be run and some followup cleanup by hand. Reboot the system and you are running on the new OS. There is a real risk in trying to keep one version for a year. I am using 4.3 which is no longer supported by the various security fixes. I try to keep to one update a year so it will be awhile before I get close again. Had to switch some hardware this year which ate up my update windows. -- -- Doug To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message
portconf port
I have been trying to figure out how to configure portconf. The 3 examples given are not much help with complex ports. I am starting with the dspam port (mail/dspam) as if I can figure that one out the rest should be easy. I first tried to use the arguments from the configure command: mail/dspam: CONFIGURE_ARGS=--with-logdir=/var/log/dspam \ --with-dspam-home=/var/db/dspam \ --with-dspam-home-owner=root \ --with-dspam-home-group=mail \ --with-dspam-home-mode=0770 \ --with-dspam-owner=root \ --with-dspam-group=mail \ --enable-homedir \ --with-storage-driver=hash_drv \ --with-delivery-agent=/usr/sbin/sendmail \ --with-dspam-mode=4511 \ --prefix=/usr/local That still brought up the options selection menu. Hitting cancel on that caused the port to start to build, but it still tried to download mysql 5.0 which I don't want. The above configure command is how I normall build dspam - in the dspam directory. Then I tried to select the options from Makefile entering the options I wanted (haven't figured out how to sent the drectories though): mail/dspam: WITH SYSLOG | DEBUG | HASH USER_HOMEDIR | SENDMAIL | SENDMAIL_LDA That skips the options selection menu fine, but still tries to download mysql 5.0 which I don't want. I then tried to add the WITHOUT options: mail/dspam: WITH SYSLOG | DEBUG | HASH USER_HOMEDIR | SENDMAIL | SENDMAIL_LDA WITHOUT DAEMON | MYSQL50 | POSTGRESQL | SQLITE3 Same results. What am I doing wrong? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Hardware Problem
I have a hardware problem with a disk drive. Last night I received millions of messages like: Oct 29 23:00:02 zook kernel: ad0: timeout waiting for write DRQ Oct 29 23:00:02 zook kernel: g_vfs_done():ad0s1e[WRITE (offset=39908835328, lengt h=8192)]error = 5 Oct 29 23:00:02 zook kernel: ad0: timeout waiting for write DRQ Oct 29 23:00:02 zook kernel: g_vfs_done():ad0s1e[WRITE (offset=39908835328, lengt h=8192)]error = 5 Oct 29 23:00:16 zook kernel: ad0: timeout waiting for write DRQ Oct 29 23:00:16 zook last message repeated 3 times Oct 29 23:00:16 zook kernel: g_vfs_done():ad0s1e[WRITE (offset=1433600, length=81 92)]error = 5 Oct 29 23:00:16 zook kernel: g_vfs_done():ad0s1e[WRITE (offset=1441792, length=81 92)]error = 5 Oct 29 23:00:16 zook kernel: g_vfs_done():ad0s1e[WRITE (offset=1449984, length=51 20)]error = 5 Oct 29 23:00:16 zook kernel: g_vfs_done():ad0s1e[WRITE(offset=8192, length=2048) ]error = 5 Oct 29 23:00:28 zook kernel: ad0: timeout waiting for write DRQ Oct 29 23:00:28 zook last message repeated 8 times They were going to messages and the console as fast as the system could log them. ad0 is not the boot disk but is used to hold large files. I have seen somewhat similar problems once before and replaced the drive. That was about a year ago. This time I am suspecting the IDE controller. Is this reasonable, or is it still likely a drive failure? Rebooting the system appears to have temporarily terminated the problem as I am no longer receiving the messages and nothing appears to have been lost on ad0. However, the bulk of data on it are the news archive and other archive files that were replaced shortly after the reboot. smartctl shows passed on the drive, but I believe I am reading that 6 sectors have been remapped. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Network Setup Question
I have a bit of an unusual network setup situation. I have a machine that is only used to store backups. It gets moved around to different locations occasionally so it has to be able to live on a 192.168.1.x or a 10.0.1.x network without reconfiguration. I also need a fixed last address byte so I can connect to it remotely. I initially set it up with DHCP and then used an alias for the .250 address on both networks. That worked, but caused problems for the local network in one location. The particular user couldn't understand why sometimes his computer got different IP addresses. So I tried to establish the 192.168.1.250 as the primary address and added an alias of 10.0.1.250. That works in both environments except that there is no default route. Is there a way to negotiate just a default route via DHCP and not an IP address? or is there a way to set the default route based on which IP address is in use? Thanks. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Network Setup Question
On Nov 10, 2006, at 19:34, Jonathan Horne wrote: On Friday 10 November 2006 19:17, Doug Hardie wrote: I have a bit of an unusual network setup situation. I have a machine that is only used to store backups. It gets moved around to different locations occasionally so it has to be able to live on a 192.168.1.x or a 10.0.1.x network without reconfiguration. I also need a fixed last address byte so I can connect to it remotely. I initially set it up with DHCP and then used an alias for the .250 address on both networks. That worked, but caused problems for the local network in one location. The particular user couldn't understand why sometimes his computer got different IP addresses. So I tried to establish the 192.168.1.250 as the primary address and added an alias of 10.0.1.250. That works in both environments except that there is no default route. Is there a way to negotiate just a default route via DHCP and not an IP address? or is there a way to set the default route based on which IP address is in use? Thanks. ___ dhclient.conf can get pretty granular as to exactly what you want from your DHCP server. myself, i use it to get everything, but to ignore the domain search mine tries to provide. man dhclient.conf and you will see tons of options (and some really good examples too). There are lots of options all right, but I couldn't find anything that would cause it not to negotiate the IP address. All of the other options are configurable. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Network Setup Question
On Nov 10, 2006, at 20:26, Lane wrote: On Friday 10 November 2006 21:56, Doug Hardie wrote: On Nov 10, 2006, at 19:34, Jonathan Horne wrote: On Friday 10 November 2006 19:17, Doug Hardie wrote: I have a bit of an unusual network setup situation. I have a machine that is only used to store backups. It gets moved around to different locations occasionally so it has to be able to live on a 192.168.1.x or a 10.0.1.x network without reconfiguration. I also need a fixed last address byte so I can connect to it remotely. I initially set it up with DHCP and then used an alias for the .250 address on both networks. That worked, but caused problems for the local network in one location. The particular user couldn't understand why sometimes his computer got different IP addresses. So I tried to establish the 192.168.1.250 as the primary address and added an alias of 10.0.1.250. That works in both environments except that there is no default route. Is there a way to negotiate just a default route via DHCP and not an IP address? or is there a way to set the default route based on which IP address is in use? Thanks. ___ dhclient.conf can get pretty granular as to exactly what you want from your DHCP server. myself, i use it to get everything, but to ignore the domain search mine tries to provide. man dhclient.conf and you will see tons of options (and some really good examples too). There are lots of options all right, but I couldn't find anything that would cause it not to negotiate the IP address. All of the other options are configurable. ___ I'm no expert, but it seems to me that your requirements are a little too optimistic. If I understand correctly, you want this machine to be able to connect to multiple heterogenous networks, and always get the same last byte for its ip. The only way to do that reliably, in my mind, is to have each dhcp server on each network assign a static address based upon the MAC address of your computer. Thats a bit much for the particular users who are housing this computer temporarily. Its bad enough that they have to put an address translation in their router to enable me to get to the .250 address. At least I can fairly easily walk them through that. If you do not have access to the DHCP server configuration on a particular network then you must manually configure the nic. That can only be done if you can access the machine which you can't in this setup since there is no default route. Assuming that you know the universe of networks that you will connect to ... say 3 or 300 possible networks ... then you could write a script in /usr/local/etc/rc.d to test various network configs ... but you might be better off just manually configuring the nic and moving on, as you cannot guarantee that the terminal byte of the ip will be available on any given network. IP just doesn't work that way.' There are only a very small number of locations for this machine, less than 5. However, its possible that at any time a new one might be necessary. This is an off-site backup machine and there needs to be someone available if we need to retrieve it. It can't be unavailable for a couple weeks. I'd be interested in any solution you may scare up, as I am faced with a similar situation. My solution is to just use static assignment, with an identifiable NETBIOS name in Samba. I am going back to the old configuration with a regular DHCP connection and then two static aliases: one for the 192 and one for the 10 addresses. That works but causes one particular user fits. I will just have to try and teach him that IP addresses will change as his DHCP reassigns them. He will have to check his computer's address and not just presume. Thanks for all the ideas. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Another Hardware Issue
I have another machine which will not boot the first time after a power on. It gives the following messages: Mounting root from ufs:/dev/ad0s1a panic: vm_fault: fault on nofault entry, addr: c6303000 syncing disks... 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 giving up on 1 buffers Uptime: 16s However, if after that you hit any key on the console or the reset button on the front panel it will boot and run just fine. This is running FreeBSD 4.6.2-RELEASE-p13. It has run fine for years until this started. There is no point to updating it as it has no users. It has no running services. It only sends a couple of status emails daily and does frequent rcp's to my production servers. Are we about to lose the motherboard? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: MAC OS X connection to FreeBSD?
On Nov 13, 2006, at 01:28, Ted Mittelstaedt wrote: Apple also doesen't use the UNIX security model. As near as I can tell their core security model is an ACL model not a user/group model. Once again this is something that's handled elsewhere. The user-group security model is alive and the heart of OS-X security. It is used throughout the system even within the user's home directory where there are files the user cannot access. This causes problems for backup progrms that want to be run by the user with a window interface as they can't backup those files. ACLs are available but not used by default. The user has to create them if desired. There used to be a FreeBSD project to add ACLs but I don't know its status. i suspect the two implementations will be very similar. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Lack of Core Dumps on signal 11
FreeBSD 6.2-RELEASE-p4 The setup I have used for earlier versions to obtain core dumps on processes that receive signals is no longer working. /var/log/ messages no longer shows core dumped for the entries and nothing appears in the core dump directory. sysctl.conf has: kern.sugid_coredump=1 kern.corefile=/usr/var/crash/%N.core /usr/var/crash has permissions of 777. There are no symlinks in the path. Symlinks didn't work in earlier versions. I need to keep all the core dumps in one directory so they can be found easily. Otherwise they end up all over the place and are qutie difficult to find. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: advice on anti-spam tools
On Apr 2, 2007, at 17:08, Kurt Buff wrote: Do you receive mail from lists such as this one? Do you receive mail from non-responding mailboxes, such as network notificationss, etc.? Do you care about your new correspondents? If you answer 'yes' to any of these messages, then a Challenge/Respons system isn't a good idea for you. TMDA is in the ports. It is a challenge/Response system that can handle non-responding mailboxes. You use one of its specially crafted addresses with those maillist servers etc. They will come through fine. There is no puzzle to solve though. The originator only has to respond. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Syslog not logging remote host
On Apr 13, 2007, at 22:44, [EMAIL PROTECTED] wrote: At 08:48 PM 4/13/2007, you wrote: "Janos Dohanics" <[EMAIL PROTECTED]> wrote: > > I'm trying capture logs from m0n0wall, but the log file is empty. > > Here is my configuration: > > On the logging machine, in /etc/rc.conf: > > syslogd_flags="-a 10.61.70.1" > > In /etc/syslog.conf: > > +10.61.70.1 > *.* /var/log/ m0n0wall.log > > /var/log/m0n0wall.log exists and writable: > > -rw-rw-r-- 1 root network 0 Apr 13 00:32 /var/log/m0n0wall.log > > The m0n0wall is configured to send logs to 10.61.70.100, which is the > logging machine. > > What am I missing? Start with tcpdump on the receiving machine: tcpdump 'port 514' to see if you're even receiving messages from the monowall machine. If not, then double-check your config on the monowall machine. If so, check the receiving machine. Bill, looks like 10.61.70.100 is receiving packets: 00:58:07.203800 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 126 00:58:33.295297 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 44 00:58:33.340779 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 49 00:59:21.436782 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 55 00:59:21.438125 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 71 00:59:21.439305 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 99 00:59:21.440458 IP gww.floco.com.syslog > 10.61.70.100.syslog: UDP, length: 92 Did you restart syslogd on both systems after making config changes? I have... Janos You might try running ktrace on the syslogd process while log messages are being sent. If you see syslogd receive the messages but not writing to a file, then there is an issue with the syslog.conf settings. It could also be logging somewhere you are not expecting. If you don't see syslogd receiving the messages then there is something blocking it or syslogd is just not listening to that host/ port. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Use of CVS
I have a medium sized application where the source is all in a CVS repository. Basically it works great as I am able to retrieve any previous version of a module when needed. Most of the changes to the application are quickly resolved, CVS committed and the production system updated in less than a day. Recently, I made a fairly large update to the application that took about 4 weeks to complete. During that time I was not able to fix small problems as there was no way to update the production system without incorporating a large number of changes from the new update that were just not working yet. Basically all small corrections were made to the new system but not incorporated into the production system until the new stuff was completed. There were no real problems from this, but it was not really convenient. Now I am going to be embarking on a revision that will take about 6 months to complete. Obviously I will not be able to wait till the completion to fix minor problems. So I am going to need to do something with branches. I have dug through the man pages and believe that is the best approach. However, given that I need to maintain the current version with a probably small number of fixes during the development process what is the best approach? Should I branch off the production version as a new branch and keep the main one for the new development or the other way around. Will it be easier to merge the fixes to the production branch back in to the new system later or should those fixes be made to both branches at the same time? Any suggestions on these approaches will be appreciated. Thanks, -- Doug ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Use of CVS
On Jan 11, 2007, at 18:28, Norberto Meijome wrote: On Thu, 11 Jan 2007 11:35:38 -0800 Doug Hardie <[EMAIL PROTECTED]> wrote: Any suggestions on these approaches will be appreciated. Thanks, I suggest you read the CVS Red book, in particular the section on branch management and merging. http://cvsbook.red-bean.com/cvsbook.html I agree with other posters, you may want to move to newer SCM systems... I've been using SVN for a while now, and couldn't be happier. There's also a SVN red book , with sections for current CVS users to understand the differences. Thanks. I have started reading them. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Loosing Ethernet Connectivity
On Jan 21, 2007, at 21:59, David Schulz wrote: hey, sure, of course i have checked the cat5 first, but it is clearly not the cable. id say it is as ted has written. what i would like to know now is how exactly happens this "hardware incompatibility"? The interface chips use a very low level "protocol" to identify the rates and modes being used by the other end. Those are dependent on voltage thresholds which sometimes are not as accurate as one would like. Components age and tolerances change which can cause the two ends to get out of sync with each other. The interface specifications also tend to change a bit over time. I don't have the exact specs for ethernet, but the same issue arose many years ago with RS-232 devices. The original specification had a threshold voltage of around 20 volts. For line drivers with 25-28 volt sources it worked great. But, 25 volts is somewhat difficult in many situations and people started fudging using 12 V sources which would work with many of the drivers that actually used a 10 V threshold. Those devices would interface with some, but not all of the older devices. Ethernet has undergone a number of changes from the original RG-8 cabling to today. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Sorta OT - Backup solutions Mac to FreeBSD
On Jan 29, 2007, at 14:00, Joe Auty wrote: I've heard of many people having problems with RsyncX and the version of rsync included in OS X crapping out and being unreliable. RsyncX and the patched rsync (the former being a GUI for the CLI rsync) that ships with OS X attempts to preserve resource forks and other file metadata (a lot of it from the OS 9 era where this stuff mattered). If you don't care at all about these attributes (I don't), I would recommend building a copy of the stock rsync from Macports, similar to FreeBSD ports in design: http://www.macports.org Here is a partial (or possibly complete) list of file metadata that I believe would be lost by using the stock rsync in OS X: - get info/Finder comments (this has been replaced with Spotlight comments in 10.4 which are saved to the Spotlight DB, not as file metadata) - application associations for files without file extensions - application associations for many OS 9 files, since OS 9 did not force file extensions and many users didn't bother with them - custom icons pasted on I have heard that also. However, I have been using it for backups for about 3 years now and every time the backup disk boots and everything I check works normally. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Quotas on 6.2
I understand quotas were broken in 6.1. I am testing 6.2 where I thought they were working again. However, it behaves considerably differently from 5.x. I set both a hard and soft limit on a user to the same value. Adding disk usage to that user past that limit succeeds. quota shows the limit as having been exceeded but with a grace period of 7 days. I don't want a grace period, but a hard limit. I used edquota -t to change the grace periods for the partition to 1 day (per the man page). However, it still shows a 7 day grace period with quota and the limit is not enforced. Did I miss something or is there still a problem with quotas. Thanks. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Quotas on 6.2
On Apr 29, 2007, at 06:26, Mark Tinguely wrote: I am using quotas in FreeBSD 6.2 with soft limits == hard limits, and the hard limits are being enforsed (mail/ftp etc). This is a non-root filesystem. There is one patch for quota that I supplied and was recently applied to -current that is appropriate for FreeBSD 5.x-7.x that has to do with file counts once the filesystem is completely full, that does not affect hard limits. I have narrowed the problem down to two items: There is a reboot called for in the handbook during the quota setup that I probably didn't do. Also, I was testing by having root copy data owned by the user with the quota. While that changed the quota used, it may not have triggered the quota check. In either case, after rebooting today and having the user do the copy, the hard quota is now enforced properly. Thanks. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Sysinstall
Some time ago I seem to remember being able to use sysinstall from a newer version CD to update an existing system. What I wanted to do was keep the original disk partitioning and just install the new system over the old one. However, with 6.2 release I don't seem to be able to do it anymore. When I get to the disk partitioning it shows the FreeBSD partition properly, but the name has changed from da0s1 to da0bs1. Hence when I get to the next step of setting up the slices it cant find da0sb1 as its not in the system. There doesn't appear to be any option to correct the name. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Sendmail ignores hosts.allow
On May 22, 2007, at 10:46, Maxim Khitrov wrote: On 5/22/07, doug <[EMAIL PROTECTED]> wrote: On Tue, 22 May 2007, Maxim Khitrov wrote: > On 5/22/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: >> I suspect sendmail is reading /etc/hosts.allow >> >> # Start by allowing everything (this prevents the rest of the file >> # from working, so remove it when you need protection). >> # The rules here work on a "First match wins" basis. >> #ALL : ALL : allow >> >> Did you comment out the above line? >> >> Steve > > Here's the entire file as it is right now: > > # Deny sendmail to all clients (temporary) > sendmail : all : deny > > # Allow anything from localhost > all : : allow > > # Process SSH deny rules > sshd : /etc/hosts.evil : deny > > # Allow everything else > all : all : allow > > Once I can get sendmail to block all connection requests, I'll move it > below the second rule. That way, only local processes will be able to > use it. For now, however, that rule is being ignored completely. The default configuration gives you what you want so I assume your goal is to see if you can make hosts.allow work within a jail. In general there are performance reasons not to use inetd to control ssh and sendmail. ssh under inetd causes more key generation. Sendmail has its own controls which give you the equivalent (or better) than can be done with inetd. I assume from an earlier post you are trying to make this work inside a jail. If thats true you must also have in the jail rc.conf inetd_flags="-wW -a your-ip-address" I assume you have this or you would not have been able to control ssh. All that said, I have only used inetd to control ftp/imap/pop3. It seems to me your specific question is: does this work inside a jail and is any special setup required to make it work with sendmail. Sorry I can not help more. Doug I'm not sure I understand what you mean... I'm not using inetd, and the default configuration doesn't block sendmail from all remote hosts. The ssh server is running all by itself, same as sendmail. The way I understand it is that as long as the server was compiled with tcp wrappers, it should follow the rules in hosts.allow. tcp wrappers must be coded into the application. The call which actually checks the access permissions in the hosts.allow file is hosts_access() (see man hosts_access). Checking through the sendmail source for version 8.13.8, there are no calls to hosts_access in the source code. You will need to patch sendmail to make it do what you want. There might be patches at www.sendmail.org for that, but I doubt it. openssh's sshd.c is probably a good template to use. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Backup advice
On May 23, 2007, at 16:27, Jason Lixfeld wrote: So I feel a need to start backing up my servers. To that end, I've decided that it's easier for me to grab an external USB drive instead of a tape. It would seem dump/restore are the tools of choice. My backup strategy is pretty much "I don't want to be screwed if my RAID goes away". That said I have a few questions along those lines: - Most articles I've read suggest a full backup, followed by incremental backups. Is there any real reason to adopt that format for a backup strategy like mine, or is it reasonable to just do a dump 0 nightly? I think the only reason to do just one full backup per 'cycle' would be to preserve system resources, as I'm sure it's fairly taxing on the system during dump 0 times. - Can dump incrementally update an existing dump, or is the idea that a dump is a closed file and nothing except restore should ever touch it? - How much does running a backup through gzip actually save? Is taxing the system to compress the dump and the extra time it takes actually worth it, assuming I have enough space on my backup drive to support a dump 0 or two? - Other folks dumping to a hard drive at night? Care to share any of your experiences/rationale? The criteria for selecting a backup approach is not the backup methodology but the restore methodology. What failures can you tollerate and what can you not afford to lose forever. Backup to a single disk leaves you with a big vulnerability if something is wrong with that backup. You stand to lose pretty much everything. If everything is stored in one location, what happens if it vanishes? My approach is dictated by the restore requirements. We have some databases that are absolutely critical. Loss of those is the end of the world. Every module that updates the database also writes a copy of the updated transaction to a log file. I rsync the log file to multiple machines separated by many miles every 10 minutes. The complete database is dumped everynight and that is also rsync'd to the same machines daily. Each of the backup machines retains several months of the full dumps and the transaction logs. From that presuming that one site remains available, I can reconstruct all but the last 10 minutes of the database. The complete system is dumped to a disk on one of the local servers weekly. A DVD is cut from that and taken off-site for retention. Actually all that is needed is the local software source and the config files as FreeBSD is easily replaced. However, since there are always times were some of the ports may not be the latest version it is easier to have the actual ones in use rather than having to checkout newer versions. The full dump is also rsync'd weekly to a couple of off-site machines. Whatever you decide to do, figure out how to recover and test it. Finding out you need something you didn't save is much less traumatic if you find out before a failure occurs. I test my restore procedures yearly. I have two machines I use at home for doing test recoveries. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
