Re: question on manpages/hier(7)
On May 9, 2013, at 10:39 AM, Dan Nelson wrote: > I don't have a /usr/local/share/man/ directory at all, and have 7300 files in > /usr/local/man/man?/ , so I'd say /usr/local/man/ is the correct location :) I wish it were that simple here. /etc/manpath.config is unmodified so I have no idea how this is getting all futzed up. I am finding files in /usr/local/share/man/man1 that were updated yesterday with others dating back to 2007. -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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"
question on manpages/hier(7)
Where should site-specific, ie local, man pages live? For instance, I have: /usr/local/man/man1/php.1.gz /usr/local/share/man/man1/php.1.gz -rw-r--r-- 1 root wheel 3597 May 6 00:38 /usr/local/man/man1/php.1.gz -rw-r--r-- 1 root wheel 3383 Dec 20 19:54 /usr/local/share/man/man1/php.1.gz My understanding is that the older one is in the right place. The newer one is registered as belonging to php5.4-14 while the old one is orphaned. I learn from lsof that the file that is actually opened and displayed is this one: /usr/local/man/cat1/php.1.gz But that's in /usr/local/man, not /usr/local/share/man. So it's in /usr/local but why not in /usr/local/share? And it's orphaned. Should it be? I have just completed a several day cleanup of my local ports installation so I'm a little mystified at this. I also rebuilt my kernel and world so I should be up-to-date there too. -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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"
securing MySQL: easiest/best ways?
Monkeying with IPv6, I discovered that globally routable addresses are what it says on the tin, so hiding behind a network appliance is not longer viable for me. An nmap scan showed the port 3306 was hanging out for all to see but I couldn't figure out how to close it off. The "--skip-networking" argument seems not to work, either in my.cnf or as an rc argument. The server just fails to start. (For some reason the socket is hard-coded to live in /tmp, regardless of what's in my.cnf but I gave up bothering about that.) What I ended up doing was adding mysql_args="--bind-address=127.0.0.1" to /etc/rc.conf. This seems to work as netstat and sockstat no longer show port 3306 listening and database connections are happening. Is this the preferred/best way? -- Paul Beard Are you trying to win an argument or solve a problem?
Re: database apps that ignore sockets? [was: Solution: mysqld fails to run, can't create/find mysql.sock]
On Jan 15, 2012, at 9:20 AM, Chuck Swiger wrote: > You're confusing two things which are different. At the risk of boring everyone on this list, I think I understand it as far as I need to: I am not the developer of the app(s) that seem to generate this issue. > If you specify a path via "--socket=/tmp/mysqld.sock", you are describing a > UNIX domain socket. While you can also specify "--host=localhost", that > would be ignored because it it implicit. If you change where the socket > lives in mysqld config or CLI options, you need to change where the clients > look for the socket as well. > > If you specify a hostname and port via "--host=localhost --port=3306", then > you are describing a TCP socket. There is no pathname involved. You could > connect regardless of where mysqld is putting the socket. If I gave the impression I didn't understand this, my mistake. The app configurations are not this granular: hostname and port are configured but there is nothing that makes clear that IF you specify localhost, you WILL BE using a domain socket which MUST BE /tmp/mysql.sock and IF you move it or your distribution prefers some other location you MAY NOT use localhost as you are now using a TCP socket which shouldn't require a hostname but because of the way the app is written, it does. Put another way, if you specify localhost, the port is ignored: I just tested this by setting the port to with a symlink to the socket placed in /tmp. It worked fine. If you change the location of the socket, you MUST use a TCP socket which mean identifying the host by name, not as localhost, even if it is localhost. There is no way to specify the location of the domain socket. It must be in /tmp. Note I am not arguing that the use of localhost requires a named domain socket, in UNIX, just that it does in this app. I learned a couple of things here. I hope I can make them clear to the people who need 'em. -- Paul Beard Are you trying to win an argument or solve a problem?
Re: database apps that ignore sockets? [was: Solution: mysqld fails to run, can't create/find mysql.sock]
On Jan 15, 2012, at 8:17 AM, Chuck Swiger wrote: > Something looking for a network location specified as a host and port (ie, > localhost:3306) is using a TCP socket. Something looking for > /tmp/mysqld.sock is using a UNIX domain socket. > > Changing the path to the UNIX domain socket will have no effect upon the port > used by the TCP socket, or vice versa. > Useful clarification but a UNIX domain socket sounds less like networking and more like interprocess communication, i.e., something explicitly tied to a single host. There is a "skip networking" option for MySQL that references the domain socket for use by processes on the same host but doesn't accept connections on port 3306. There's no indication that using localhost will default to a domain socket which will explicitly be looked for in /tmp and if you put it anywhere else, you must specify a hostname to access the TCP socket. I'll quote your definition in the bug report as it seems crystal clear. -- Paul Beard Are you trying to win an argument or solve a problem?
database apps that ignore sockets? [was: Solution: mysqld fails to run, can't create/find mysql.sock]
On Jan 14, 2012, at 11:15 AM, Chuck Swiger wrote: > Anyway, doesn't the mysql port want to keep the socket under > /var/run/mysql/mysqld.sock or some such, to avoid issues with /tmp? Turns out some applications won't work if you move the socket if they are configured to access localhost. Seems like a misunderstanding of networking if you can specify a port number in a configuration file but the application looks to the filesystem for the socket. There is no way to specify a file location so it seems doomed to fail — as it did. The apps in question are net-mgmt/cacti and net-mgmt/cacti-spine. -- Paul Beard Are you trying to win an argument or solve a problem?
Re: Solution: mysqld fails to run, can't create/find mysql.sock
On Jan 14, 2012, at 11:15 AM, Chuck Swiger wrote: > The meaning seems obvious enough; mysqld was unable to bind to the socket, > which is what perror() meant with "Permission denied": Really? I read this: > 120114 9:39:04 [ERROR] Do you already have another mysqld server running on > socket: /tmp/mysql.sock ? as "there is an existing socket that seems to be in use: what's up with that?" The message references a file that does not exist (but that mysql will cheerfully remove if found). There was no existing socket. Those two lines, taken together, tell me that a. mysql can't run without a socket and b. it thinks another process is running, bound to a socket that doesn't exist. Clear as mud. How about [ERROR] socket: /tmp/mysql.sock not found and/or [ERROR] socket:/tmp/mysql.sock could not be created perhaps with a helpful hint about permissions. If this was unusual, that would be one thing but I found quite a few references to the problem before I found the solution. Maybe it's a housekeeping thing but why would mysql need to destroy the file it uses for a socket and then recreate it when it could simply examine it and reuse it? > Anyway, doesn't the mysql port want to keep the socket under > /var/run/mysql/mysqld.sock or some such, to avoid issues with /tmp? Apparently not, as I commented out any reference to it in my.cnf and still saw the same messages about /tmp/mysql.sock. It seems to work if spelled out explicitly. -- Paul Beard Are you trying to win an argument or solve a problem?
Solution: mysqld fails to run, can't create/find mysql.sock
Woke up to a screenful of error messages about failed mysql backups and found that for some reason, mysql was refusing to run at all. The issue was not just a missing mysql.sock but an inability to create one. I could do it by hand or at least create a file with the same name and permissions but it would be removed on the next attempt and then not replaced. Turns out the permissions on /tmp were not right. I didn't note them beforehand but setting them 1777 solved it. I would be interested in knowing how those permissions got changed. I rebooted the system early on in the process as I kept seeing messages like this: 120114 9:39:04 [ERROR] Can't start server : Bind on unix socket: Permission denied 120114 9:39:04 [ERROR] Do you already have another mysqld server running on socket: /tmp/mysql.sock ? Those are rubbish as error messages as they don't say the file can't be created or give any indication of the actual problem. This is all more a problem for the mysql developers than FreeBSD but I am posting it to the list in case anyone else gets bitten by it. -- Paul Beard Are you trying to win an argument or solve a problem?
Re: wireless access point in FreeBSD 8.2p2
On Aug 28, 2011, at 5:28 PM, Warren Block wrote: > IMO, the wireless section is already so stuffed full of detail that it > obscures the basics. In fairness, it's a complicated topic. But I'd much > rather see a simple setup for the 80% use case followed by another section > with all the grimy details that most people won't need. That's why I did > this: http://www.wonkity.com/~wblock/docs/html/wireless.html I'll read through that and see if it helps me at all. I found this — "'Hidden' SSIDs are not really hidden. They make network setup more difficult and provide no real security benefits." — interesting. I assume you could figure them out from wardriving? -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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: wireless access point in FreeBSD 8.2p2
On Aug 28, 2011, at 3:17 PM, Warren Block wrote: > In 8.x, a virtual wlan0 device is created to speak to the actual device, ath0 > in this case. It's normal. Maybe I'm just confused by normality. I guess what would help in the Handbook, if nowhere else, is the *full* output of ifconfig(8) for purposes of comparison and elucidation. I see it called with just one interface but unless you're on the console, it's likely you're adding the wlan interface to an existing wired interface. And at the risk of stating the obvious, there has to be a wired interface for it to actually work as an access point, no? -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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"
Closed [was Re: wireless access point in FreeBSD 8.2p2]
On Aug 28, 2011, at 12:38 PM, Bill Tillman wrote: > I did this to have the experience with it and to have a backup to my Netgear > wireless router. The trouble was the Netgear wireless AP device works so well > and is plenty fast, unlike what I was getting with my FreeBSD server. The > Netgear device has been working 24/7 for almost 2 years now so I just gave up > on the FreeBSD option. I think you sold me on the futility of this exercise. My Linksys has been in service longer than that, 5 years at least. Maybe one of the micro-flavors of FreeBSD or NetBSD might be better suited to this, on a low-powered disused hardware platform. It doesn't seem reliable enough to spend any more time on. If I could make it work, I could like to compare what it took to the various approaches I have read (including in the Handbook which is usually very dependable: I used it earlier today to refresh my memory on adding a swapfile). I'm calling it closed for now. Better use of my time to just find a backup AP. -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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: wireless access point in FreeBSD 8.2p2
On Aug 28, 2011, at 7:04 AM, CyberLeo Kitsana wrote: > It is especially useful when you cannot ping, as it can tell you if the > packets are even arriving. The "no route to host" result makes me think the packets aren't going far ;-) The new device and the wired interface are at adjacent numeric addresses and all the devices here are in the same subnet behind the WRT54G and that is behind the cable co's black box. I think I may be more confused now than when I started. One thing that has seemed opaque to me is that both ath0 and wlan0 display when I run ifconfig and look very similar: makes me think they might be stepping on each other. Or it's just one more thing I don't understand :-( ath0: flags=8843 metric 0 mtu 2290 ether 00:0d:88:93:21:3a media: IEEE 802.11 Wireless Ethernet autoselect mode 11g status: running wlan0: flags=8843 metric 0 mtu 1500 ether 00:0d:88:93:21:3a inet 192.168.0.26 netmask 0xff00 broadcast 192.168.0.255 inet6 fe80::20d:88ff:fe93:213a%wlan0 prefixlen 64 scopeid 0x8 nd6 options=3 media: IEEE 802.11 Wireless Ethernet autoselect mode 11g status: running ssid lower channel 8 (2447 MHz 11g) bssid 00:0d:88:93:21:3a regdomain FCC indoor ecm authmode OPEN privacy OFF txpower 27 scanvalid 60 protmode CTS wme burst dtimperiod 1 -dfs I know (or think I do) that ath0 is the real interface and wlan0 is a virtualized or cloned or something handle to it. But the similarities (both are running, both show the same info for media) trouble me. The only thing that makes me think I'm doing anything here is that wlan0 is actually assigned to channel 8. I can sort of see that getting it working as a client would be instructive and I think I did that some time ago (perhaps in 7.x) but since you reuse almost nothing but the hardware, I don't see a lot of value in that, other than verifying that the hardware works and that you can follow the instructions. The latter can be a challenge, I'll admit. So to recap: the idea of this was to provide a redundant spare for the WRT54G, behind a cable modem, in a private network, with the only security being at the AP • No ipfw or any of that, as it wouldn't be visible on the public internet. • I'll add WPA/2 once it works (that seems trivial, as I have been able to authenticate to the AP even though it didn't pass any packets beyond that). • It would deal with static addresses (I could add dhcp later, once this was working, as phones and other devices are more easily dealt with that way). So it looks like a bridge, if it joins an Ethernet network and an 802.11-based one. Curiously, none of the instructions I have seen mention bridging, even though the explicitly connect Ethernet and wireless. And all the HOWTOs look simple, the work of a few minutes of copy and paste. I think I may just shelve this and if needed, turn up my Time Capsule's wireless capability (if it would play nicely and extend the WRT54G, I'd be using it now). And APs that support open source firmware are not that hard to find, though Tomato doesn't support as many as the *-wrt variants. *grumble* -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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: bridged wireless access point in FreeBSD 8.2p2
On Aug 27, 2011, at 8:48 PM, CyberLeo Kitsana wrote: > This looks correct so far, for an unsecured wireless access point. The > wlan0 device is the interface you will use for communicating; the ath0 > device exists solely as a target for wlan0 creation. After some more head scratching, it sounds like what I want is a bridge. Reading if_bridge(4), the first example looks a lot like what I am trying to do. My initial attempt didn't fare any better but I expect I have a lot of cruft to dig through from the earlier attempts. Did I misread this? Does sending packets between two physical interfaces require a bridge? No other docs have mentioned it but of their "easy instructions" have worked either. -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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: wireless access point in FreeBSD 8.2p2
On Aug 27, 2011, at 8:48 PM, CyberLeo Kitsana wrote: > tcpdump(1) is your friend; it seems cryptic and obtuse at first glance, > but it will help immensely > I wasn't sure there was any reason to use that yet: I can't even ping it from another host. > wlan0 itself will not assign v4 addresses to clients; you need a DHCP > server for that > I plan to use static addresses as I do already (this is just a backup in case my WRT54G develops any issues). > The hostap machine must be explicitly told to route packets, by setting > gateway_enable="YES" in rc.conf and adding the appropriate routes > I have that and the existing wired interface has the route set (I am connecting through that to make this work). This raises the question of whether I am expecting the functionality of a bridge without having specifically made one. > If you're intending this to be a home gateway, you will likely also need > NAT. I think NAT is handled by the telco hardware (on cable) for now. Hmm, starting to think this may not work as I expect. It might be fine as an additional AP but not as a replacement without some configuration changes that I will have forgotten how to make by then. The WRT box runs the PPPoE connection for DSL which I should be switching back to. I'm sure it can be done with this but I think I'm asking for trouble. So maybe this is a solution in search of a problem. Might be to just find a spare WRT54G or its modern equivalent. But that doesn't mean I don't want to figure this out. -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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"
wireless access point in FreeBSD 8.2p2
I seem to be missing something, possibly from reading too many HOWTOs. What I am trying to do is get a system with a wireless card to stand in as a wireless AP should my aging LinkSys base station develop a tragic smoke leak. It's an ath0-based card and the following steps suggest it should work (it has HOSTAP capabilities and offering robust encryption). ifconfig wlan0 create wlandev ath0 ifconfig wlan0 list caps drivercaps=6f85ed41 cryptocaps=f But various permutations of rc.conf, hostap.conf and many iterations of /etc/rc.c/netif restart leave me with two ifconfig entries, one of the ath0 interface and one for wlan0. None of the examples show this so I suspect it's wrong. The IP address is pingable from the host it's installed in but not from anywhere else. And I can see the AP from another system and attach to it but it doesn't route any traffic. from /etc/rc.conf: wlans_ath0="wlan0" create_args_wlan0="wlanmode hostap" ifconfig_wlan0="inet 192.168.0.26 netmask 255.255.255.0 ssid lower mode 11g channel 8" from hostap.conf: interface=wlan0 debug=1 ctrl_interface=/var/run/hostapd ctrl_interface_group=wheel ssid=lower wpa=0 redacted results of ifconfig: ath0: flags=8843 metric 0 mtu 2290 ether 00:0d:88:93:21:3a media: IEEE 802.11 Wireless Ethernet autoselect mode 11g status: running wlan0: flags=8843 metric 0 mtu 1500 ether 00:0d:88:93:21:3a inet 192.168.0.26 netmask 0xff00 broadcast 192.168.0.255 inet6 fe80::20d:88ff:fe93:213a%wlan0 prefixlen 64 scopeid 0x7 nd6 options=3 media: IEEE 802.11 Wireless Ethernet autoselect mode 11g status: running ssid lower channel 8 (2447 MHz 11g) bssid 00:0d:88:93:21:3a regdomain FCC indoor ecm authmode AUTO privacy OFF txpower 27 scanvalid 60 protmode CTS wme burst dtimperiod 1 -dfs Yes, I am trying it without any encryption until I see some packets being passed. It seems like a lot of people are getting this to work but I'm not able to follow how they did it. -- Paul Beard Are you trying to win an argument or solve a problem? ___ 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"
screen and curses(?) output
I have discovered the joys of screen to manage processes between/across logins, but I have one small problem. If I run portupgrade and a dialog is displayed, I can't choose any of the displayed options. A tab will move through them but a space bar or arrow key is the same as hitting OK with no changes. is there some setting or different termcap I should use? this is what stty -e displays: speed 9600 baud; 40 rows; 80 columns; lflags: icanon isig iexten echo echoe -echok echoke -echonl echoctl -echoprt -altwerase -noflsh -tostop -flusho pendin -nokerninfo -extproc iflags: -istrip icrnl -inlcr -igncr ixon -ixoff ixany imaxbel -ignbrk brkint -inpck -ignpar -parmrk oflags: opost onlcr -ocrnl -oxtabs -onocr -onlret cflags: cread cs8 -parenb -parodd hupcl -clocal -cstopb -crtscts -dsrflow -dtrflow -mdmbuf discard dsusp eof eol eol2erase erase2 intrkill ^O ^Y ^D^? ^H ^C ^U lnext min quitreprint start status stopsusptime ^V 1 ^\ ^R ^Q ^T ^S ^Z 0 werase ^W -- Paul Beard / www.paulbeard.org/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: kernel debugging: seeing swap issues in 7.0
On Sat, May 24, 2008 at 1:28 AM, Kris Kennaway <[EMAIL PROTECTED]> wrote: > > There is no bug in gcc. Your system does not have enough memory to > run the compilation job, so you are thrashing it severely. Try > turning down the CFLAGS optimization level, or use a precompiled > package from the FTP site or built on another machine. That's as may be, but the initial warning/error message was that the gcc folks wanted a big filed. If there is a package for this, I'll go that route. -- Paul Beard / www.paulbeard.org/ <[EMAIL PROTECTED]/[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: kernel debugging: seeing swap issues in 7.0
On Fri, May 23, 2008 at 2:08 PM, Kris Kennaway <[EMAIL PROTECTED]> wrote: > > You can try increasing the timeout by editing the relevant kernel > source, but if it's failing to reply to the I/O after 30 seconds then > something is drastically overloaded on your system. Well, not much is running, and I have recently built a new kernel and world in less time than it takes to get this one port upgraded. I think there is something else amiss. Should this really be soaking up 100Mb of RAM? PID USERNAME THR PRI NICE SIZERES STATETIME WCPU COMMAND 32272 root 1 -200 101M 70340K swread 3:04 0.93% cc1 I have filed a bug against gcc, per instructions. There seems to be an issue somewhere in py-gtk that is exacerbated by the low resources on this system. Thanks. -- Paul Beard / www.paulbeard.org/ <[EMAIL PROTECTED]/[EMAIL PROTECTED]> ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
kernel debugging: seeing swap issues in 7.0
I see a lot of these as my system grinds to a halt. It never crashes (I reboot it when it gets really boggy . . . May 23 11:41:13 stinky kernel: swap_pager: indefinite wait buffer: bufobj: 0, blkno: 49155, size: 4096 May 23 11:41:21 stinky kernel: swap_pager: indefinite wait buffer: bufobj: 0, blkno: 48905, size: 4096 This is the top of the kernel debug Unread portion of the kernel message buffer: <118>May 23 12:48:46 stinky syslogd: exiting on signal 15 Waiting (max 60 seconds) for system process `vnlru' to stop...done Waiting (max 60 seconds) for system process `bufdaemon' to stop...done Waiting (max 60 seconds) for system process `syncer' to stop... Syncing disks, vnodes remaining...1 0 0 done All buffers synced. swap_pager: I/O error - pagein failed; blkno 35668,size 4096, error 5 panic: swap_pager_force_pagein: read from swap failed Uptime: 14h20m56s Physical memory: 115 MB Dumping 37 MB: 22 6 #0 doadump () at pcpu.h:195 195 __asm __volatile("movl %%fs:0,%0" : "=r" (td)); I have partition-based swap and swap files I set up after gcc was running out of memory. Device: 1048576-blocks Used: /dev/ad0s2b 139 0 /dev/md0 128 0 Any ideas what I can do (besides buy more hardware)? -- Paul Beard / www.paulbeard.org/ <[EMAIL PROTECTED]/[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: what is pkg_create doing?
On May 31, 2007, at 11:22 AM, Conrad J. Sabatier wrote: No, by default, portupgrade runs pkg_create *before* installing the newly built port, to create a backup of the old version in case something goes wrong. Depending on the size of the old port (package), this can take an appreciable amount of time, even on a fast system. Packages are only built *after* installation if one explicitly tells portupgrade to do so, via the "-p" or "--package" switches. This is consistent with what I am observing, as the command output claims the package is installed, I then see a delay (minutes? on a dual core 2.4GHz system?) while pkg_create runs, and then cleanup happens. The man page lists several package-related ENVIRONMENT variables, which may or may not provide a means to disable some or all of this package creation; I don't know for sure, as I've never tried changing or unsetting them to see what may happen. If you're curious, though, it may be worth experimenting with, although I would certainly advise against disabling the precautionary backup package creation before the new port is successfully installed. Thanks. I'll take a look there. I wasn't sure if something changes in how ports are done in the Moderne Age. -- Paul Beard words: http://paulbeard.org/wordpress pictures: http://www.flickr.com/photos/pdb206/ Are you trying to win an argument or solve a problem? PGP.sig Description: This is a digitally signed message part
what is pkg_create doing?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I realize it should be obvious from the name but it seems to linger for several minutes after each package is installed. PID USERNAMETHR PRI NICE SIZERES STATE C TIME WCPU COMMAND 55763 root 11210 1852K 1232K RUN 0 2:38 95.73% pkg_create I am new to 6.2 (being a belated migrator from 4.x): is there a knob somewhere to turn off pkg creation? Not that I see any packages being kept anywhere . . . . I'm not subscribed, so CCing would ensure I get your reply. - -- Paul Beard words: http://paulbeard.org/wordpress pictures: http://www.flickr.com/photos/pdb206/ Are you trying to win an argument or solve a problem? -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.7 (Darwin) iD8DBQFGXO1AjE2ksZfa4ZURAlIXAJsE844ANT3n0szi55MvP1w5+cAsxwCgzvxg RVB/DwVspsN24/UqdPzRCPE= =Vaao -END PGP SIGNATURE- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
verbose sysctl messages in dmesg?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 for some reason, my dmesg output is filled with what looked sysctl options: [/usr/home/paul]:: dmesg 14 fastforwarding RW *Handler Int 15 keepfaith RW *Handler Int 100 subnets_are_local RW *Handler Int 101 fw RW Node 100 enable RW *Handler Int 101 one_pass RW *Handler Int 102 debug RW *Handler Int 103 verbose RW *Handler Int 104 verbose_limit RW *Handler Int 105 dyn_buckets RW *Handler Int 106 curr_dyn_buckets R *Handler Int 107 dyn_count R *Handler Int 108 dyn_max RW *Handler Int 109 static_count R *Handler Int 110 dyn_ack_lifetime RW *Handler Int 111 dyn_syn_lifetime RW *Handler Int 112 dyn_fin_lifetime RW *Handler Int 113 dyn_rst_lifetime RW *Handler Int 114 dyn_udp_lifetime RW *Handler Int 115 dyn_short_lifetime RW *Handler Int 116 dyn_grace_time R *Handler Int 102 maxfragpackets RW *Handler Int 103 maxfragsperpacket RW *Handler Int 104 sendsourcequench RW *Handler Int I get almost 1000 lines of this stuff, to the exclusion of anything else in dmesg. Is this something I can toggle off? This is in FreeBSD 4.11, built from sources pulled just a couple of days ago. - -- Paul Beard contact info: www.paulbeard.org/paulbeard.vcf Are you trying to win an argument or solve a problem? -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (Darwin) iD8DBQFDTd68fHLPwpj1/JQRAh8rAKCAjNn4uNgoaMSUammdhof7OynsHACgmfY0 ChPiPdkP/UW2fcHNie+ckqg= =kp0r -END PGP SIGNATURE- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Still have questions on portupgrade database problems
I am still seeing this error if I run portupgrade -a: ---> Session started at: Tue, 19 Jul 2005 08:30:17 -0700 [Rebuilding the pkgdb in /var/db/pkg ... - 542 packages found (-0 +542) .. done] ---> Session ended at: Tue, 19 Jul 2005 08:34:00 -0700 (consumed 00:03:42) /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:322:in `deorigin': cannot convert nil into String (PkgDB::DBError) from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:915:in `tsort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:914:in `each' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:914:in `tsort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:906:in `each' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:906:in `tsort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:928:in `sort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:932:in `sort_build!' from /usr/local/sbin/portupgrade:674:in `main' from /usr/local/sbin/portupgrade:207:in `initialize' from /usr/local/sbin/portupgrade:207:in `new' from /usr/local/sbin/portupgrade:207:in `main' from /usr/local/sbin/portupgrade:1869 I have heeded the advice in UPDATING: adjust the PORTS_DBDRIVER variable (in your environment or in pkgtools.conf) to either bdb1_hash or dbm_hash. but it doesn't seem to take. I have bdb1_hash in pkgtools.conf. I have even tried -q/--noconfig to only take environment variables. [/var/db/pkg]# export PORTS_DBDRIVER=bdb1_hash; pkgdb -qF ---> Checking the package registry database [Rebuilding the pkgdb in /var/db/pkg ... - 542 packages found (-0 +542) .. what am I missing? FWIW, I can use an incantation of portmanager to find the outdated ports and use portupgrade on the one by one. -- Paul Beard contact info: www.paulbeard.org/paulbeard.vcf Are you trying to win an argument or solve a problem? PGP.sig Description: This is a digitally signed message part
Re: portupgrade error [cannot convert nil into String (PkgDB::DBError)]
On Jul 9, 2005, at 6:01 AM, Kris Kennaway wrote: Did you read /usr/ports/UPDATING? The last change I found mentioned the use of libc and the PORTSDB_DRIVER. I made that change ages ago and have tried the other variants to get around this. -- Paul Beard contact info: www.paulbeard.org/paulbeard.vcf Are you trying to win an argument or solve a problem? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
portupgrade error [cannot convert nil into String (PkgDB::DBError)]
I am having some problems with portupgrade. The error message is "/ usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:322:in `deorigin': " (more below). I see there have been isolated occurrences of this in June of 2003 and 2004, curiously, but I haven't found the solutions there to work. Removing the ports tree and re-fetching it, removing and rebuilding pkgdb, removing and reinstalling portupgrade/ruby, etc, even building a new kernel and world, seem ineffective. I'm still on 4.x (FreeBSD red.paulbeard.org 4.11-RELEASE-p11 FreeBSD 4.11-RELEASE-p11 #0: Tue Jul 5 10:46:00 PDT 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/RED i386) if that helps. Please include me on replies as I am off the list. Thanks for any pointers. [/]# portsdb -Ufu Updating the ports index ... Generating INDEX.tmp - please wait..Warning: Duplicate INDEX entry: freeciv-gtk2-2.0.1_2 Warning: Duplicate INDEX entry: mod_frontpage2-5.0.2.2635 Warning: Duplicate INDEX entry: mod_jk2-apache2-2.0.2 Warning: Duplicate INDEX entry: mod_rpaf-ap2-0.5 Done. done [Updating the portsdb in /opt/ports ... - 13138 port entries found . 1000.2000.3000.4000.5000.6000... .. 7000.8000.9000.1.11000.12000 .13000. . done] ([EMAIL PROTECTED])-(03:34 PM / Fri Jul 08) [/]# portupgrade -avvv ---> Session started at: Fri, 08 Jul 2005 15:37:34 -0700 ---> Session ended at: Fri, 08 Jul 2005 15:37:56 -0700 (consumed 00:00:22) /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:322:in `deorigin': from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:915:in `tsort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:914:in `each' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:914:in `tsort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:906:in `each' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:906:in `tsort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:928:in `sort_build' from /usr/local/lib/ruby/site_ruby/1.8/pkgdb.rb:932:in `sort_build!' from /usr/local/sbin/portupgrade:674:in `main' from /usr/local/sbin/portupgrade:207:in `initialize' from /usr/local/sbin/portupgrade:207:in `new' from /usr/local/sbin/portupgrade:207:in `main' from /usr/local/sbin/portupgrade:1869 -- Paul Beard contact info: www.paulbeard.org/paulbeard.vcf Are you trying to win an argument or solve a problem? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
one for the archives
I had posted a couple of questions about my system suddenly refusing to build shared libraries and making my ports tool-based builds fail. The symptoms were that the host type was being misread as kfreebsd/gnu suggesting I had some Debian/FreeBSD frankenhost, and shared libraries would fail to build on most ports. With some research from Pav Lucistnik <[EMAIL PROTECTED]>, a rogue file -- /usr/include/features.h -- was discovered and removed. It's dated Nov 21, 2001, and is claimed by no package so I have no idea how it got there. But as far as I can tell, that was the culprit. I will add this info to the PR as well. -- Paul Beard paulbeard [at] mac.com
lingering problems with ports collection
I have been having (and reporting) some problems with my ports collection and I can't seem to get them resolved. For some reason, the system is rejecting the ports collection like a mismatched organ. http://lists.freebsd.org/pipermail/freebsd-questions/2004-March/ 040320.html http://lists.freebsd.org/pipermail/freebsd-questions/2004-March/ thread.html#38348 The upshot is that I can't reliably build ports due to some problems with shared libs not getting built. Configure doesn't see this as a freeBSD system, but as some hybrid GNU/FreeBSD beast and sometimes sets the host type as unknown-kfreebsd-GNU. Then I find that while a port is recorded as installed, ports that depend on its shlibs fail. I can test this by making a package when I make the port, but the port is installed before packaging, so the end result is that I may have a package recorded as installed but I know i can't rely on it. (that seems backwards to me: what if I only want to build packages w/o installing them? can i do that? To resolve this, I have: built a new kernel and world from fresh sources (in the process finding another problem to do with /usr/src/gnu/usr.bin/perl needing to rebuilt). tossed my whole ports directory and moved aside the pkg database, pulling a fresh tree from CVS. Still I see these problems. Also, dependencies don't get built automatically: if I build sysutils/portupgrade, I don't get prompted to build security/openssl or even lang/ruby: pkgdb will tell me, but not the ports tools themselves. What sometimes helps is to change the USE_LIBTOOL directive from 13 and 15: libtool 1.3 seems to be part of my problem. But that's not always working. To be clear, I can build and install just from the source directory (/usr/ports/{PORTNAME}/work/portname). But that doesn't get it registered and if I got back up to the port's directory, the make commands will kick back errors I didn't see in the source directory. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
port installation problems progress
I had been having some problems installing ports: they manifested as configure misreading the system type and setting up libtool so it wouldn't build shared libs. After some correspondence with a FreeBSD team member and a port maintainer, I decided to start afresh with a kernel and world reinstall. This led me to the solution. The email excerpted below mentioned almost the same error I was seeing: http://www.atm.tut.fi/list-archive/freebsd-stable/msg04153.html For the last week or so, I've been having trouble building kernels against a RELENG_4 source tree. A 'make buildkernel' from /usr/src fails as follows: perl5 /usr/src/sys/kern/vnode_if.pl -h /usr/src/sys/kern/vnode_if.src syntax error at /usr/src/sys/kern/vnode_if.pl line 135, near "{}" Execution of /usr/src/sys/kern/vnode_if.pl aborted due to compilation errors. I took the advice suggested below. http://www.atm.tut.fi/list-archive/freebsd-stable/msg04174.html I believe I have these resolved. If anybody's interested, it took a reboot, buildworld followed by an install from /usr/src/gnu/usr.bin/perl, then buildkernel went fine. At Mike Tansca's suggestion, I'd tried a buildworld (against a fresh source tree), and I was getting cc errors - the reboot appears to have fixed those. My build machine is now happily running a new kernel/world. Some packages are building OK but I still see issues on some ports. Any ideas how this might have gotten into this state and how I can resolve it completely? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
libtool/shared libraries problem
In the course of cleaning up this b*rked port installed, I found that popt wouldn't install properly. It would report no errors but ports that depended on it would bail out, unable to find shlibs. configure, running under the ports system, returns this: creating libtool checking host system type... i386-unknown-kfreebsd4.9-gnu <-- what's that all about? . . . . . checking dynamic linker characteristics... no checking if libtool supports shared libraries... no checking whether to build shared libraries... no checking whether to build static libraries... yes . . . . Cleaning out the work directory, doing a make deinstall in libtool13 gives me this more reasonable outcome. checking host system type... i386-unknown-freebsd4.9 . . . . checking if libtool supports shared libraries... yes The acid test is to see if "make package" will work since it seems to check the created files against a list that "make install" doesn't seem to do. I just did a portinstall rpm (where I was stuck with popt being broken) and it installed just fine. It did however kick back the same warning I have seen previously that libtool13 is deprecated and libtool15 is preferred. There is no warning about the popt port in UPDATING. -- Paul Beard paulbeard [at] mac.com
Re: workaround for the expat problem
On Apr 1, 2004, at 12:05 PM, Kevin D. Kinsey, DaleCo, S.P. wrote: Users of expat2 (and its many dependencies) should do the following to properly update expat2 and all of its dependencies: portupgrade -rf textproc/expat2 I wish I had kept a record of how many times I ran that and had it fail. The problem was that for some reason the expat2 port was unable to create and install shared libraries due to something that was getting picked up in the configure run. As noted in my note about the workaround, running the build without using the ports system infrastructure allowed the shlibs to be created and install and then -- and only then -- could the dependent ports find the expat.5 library and complete their upgrades. -- Paul Beard paulbeard [at] mac.com
Re: some anomalies in my system?
On Mar 31, 2004, at 9:21 PM, Kris Kennaway wrote: OK, feel free to ignore my advice as you see fit. Looking back through the thread, you asked if I had installed any non-ports software or some other variant of libtool: the answer was no. I didn't see any advice, so I took the initiative to find out what libtool might be doing. I did discover that there was a dependency on a deprecated version of libtool (1.3 where 1.5 seems to be the current/preferred version). And I was able to get around my problem by working from the distfiles and leaving out whatever the ports system was doing (that was where the inability to build shlibs was getting in the mix). As noted in my followup post to the list, once I installed from the {$PORTDIR}/work/ directory, all went well and I have now resolved my problem. Apologies if I offended. The insight on libtool is appreciated. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 31, 2004, at 9:21 PM, Kris Kennaway wrote: You want the packages-4-stable directory; RELEASE packages are not updated. See http://www.freebsd.org/ports for more details. Ah, I see. There was an earlier email to the effect that the packages I needed were under 4.9-RELEASE. -- Paul Beard paulbeard [at] mac.com
workaround for the expat problem
I removed all the versions of libtool, and tried rebuilding expat but that didn't help. It would install libtool13 and I would get funky configure output and a failed install. I then went into ${PORTSDIR}/textproc/expat2/work, removed the expat directory, re-extracted from the distfile, and ran configure, make and name install from there. That seemed to be OK, then I went back to ${PORTSDIR}/textproc/expat2/ and did a make install WITH_PKG_REGISTER=1. That worked well enough to build gettext 0.13 with its dependency on libiconv. So all looks to be normal, or close to it. If anyone has any ideas what this is all about and how best to cope with it going forward, I'd be grateful to know about it. Thanks. -- Paul Beard paulbeard [at] mac.com
Re: some anomalies in my system?
On Mar 31, 2004, at 8:48 PM, Kris Kennaway wrote: I took that as a hint to look at the versions of libtool: there wasn't a version I didn't have installed, so I removed them all and decided to let them get handled as dependencies. That's not what I asked. I realize that, but that was the first step to getting this fixed or at least worked around. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 31, 2004, at 7:52 PM, Kris Kennaway wrote: You're installing out of date packages then. The ones on the ftp site are current and provide libexpat.so.5. This is what's available for 4.9-RELEASE: [/opt/ports/packages/All]# pkg_info -L expat-1.95.6_1.tgz Information for expat-1.95.6_1.tgz: Files: /usr/local/man/man1/xmlwf.1.gz /usr/local/bin/xmlwf /usr/local/include/expat.h /usr/local/lib/libexpat.a /usr/local/lib/libexpat.so /usr/local/lib/libexpat.so.4 Likewise 5.2-RELEASE: I just pulled that package and checked. The Makefile in my ports tree has this: PORTVERSION=1.95.7. this would/should give me expat.5. -- Paul Beard paulbeard [at] mac.com
Re: some anomalies in my system?
On Mar 31, 2004, at 7:58 PM, Kris Kennaway wrote: It still looks like you have a bogus non-ports version of libtool installed that is providing the wrong information to the configure script. I took that as a hint to look at the versions of libtool: there wasn't a version I didn't have installed, so I removed them all and decided to let them get handled as dependencies. Expat seems to be relying on libtool13: ===> Patching for expat-1.95.7 ===> Applying FreeBSD patches for expat-1.95.7 ===> expat-1.95.7 depends on file: /usr/local/libexec/libtool13/libtool - not found ===>Verifying install for /usr/local/libexec/libtool13/libtool in /usr/ports/devel/libtool13 ===> NOTICE: This port is deprecated; you may wish to reconsider installing it: Please use devel/libtool15 instead. It is scheduled to be removed 31st December 2004. -- Paul Beard paulbeard [at] mac.com
some anomalies in my system?
Working through this issue with expat not building shared libs, the maintainer noticed this: creating libtool checking host system type... i386-unknown-kfreebsd4.9-gnu checking whether to build shared libraries... no checking whether to build static libraries... yes what isn't letting us build shared libs and what's that funky system type? Other ports don't have this issue, at least from the others I chose at random. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 31, 2004, at 5:38 PM, Kris Kennaway wrote: Well, then you can't use it :) There are certainly up-to-date packages for the basic ports like expat that you're having problems with. So here's my problem. I install expat and gettext from packages. Then I install *anything* that depends on those, and they need to be rebuilt. The expat build fails. The gettext fail as a result, and then all the rest do as well. I'm not sure how up to date they are if the ports that depend on them want to rebuild them. Information for expat-1.95.6_1.tgz: Files: /usr/local/man/man1/xmlwf.1.gz /usr/local/bin/xmlwf /usr/local/include/expat.h /usr/local/lib/libexpat.a /usr/local/lib/libexpat.so /usr/local/lib/libexpat.so.4 gettext 0.13.x wants expat.5, so that's not as up to date as it might be. I've dropped a note to the maintainer: perhaps there's something he can point out. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to makestuff work?)
On Mar 31, 2004, at 4:59 PM, Matt Emmerton wrote: How is 4.9 ancient? It's the most recent release supported for production use. (Refer to http://www.freebsd.org if you doubt this.) That was a joke ;-) You can find all the packages you want for 4.9 at ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/4.9-RELEASE/packages Thanks. I'll see if those will work. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 31, 2004, at 4:29 PM, Kris Kennaway wrote: You're not doing it right, then :-) Use portupgrade with the -PP switch to force the use of packages. What if there isn't a package for a given port? I wasn't aware there were packages (though I suppose for an ancient release like 4.9 there might be). -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 31, 2004, at 3:28 PM, Kris Kennaway wrote: What architecture is this on? Some people have reported problems on amd64 with shared libraries not being created - this appears to be something to do with libtool, but I don't see it on my amd64 box. It's on x86, running 4.9. To work around mysterious port problems you can always just install the packages instead. I tried that, but some ports want to rebuild their dependencies and I'm back where I started. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 31, 2004, at 2:29 PM, Charles Swiger wrote: Stop using portupgrade for a moment; do a "make deinstall" of the expat port (or use "pkg_delete -f"), do a "make clean", and then do a "make reinstall". been there, done that, got the bloody knuckles to show for it. [/usr/ports/textproc/expat2]# pkg_info -g /var/db/pkg/expat-1.95.7/ Information for expat-1.95.7: Mismatched Checksums: pkg_info: /usr/local/lib/libexpat.so doesn't exist pkg_info: /usr/local/lib/libexpat.so.5 doesn't exist I think something's broken. -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
Well, after a lot of different attempts to get past this problem, it seems that expat isn't building all the files it needs to. ===> Building package for expat-1.95.7 Creating package /usr/ports/packages/All/expat-1.95.7.tgz Registering depends:. Creating gzip'd tar ball in '/usr/ports/packages/All/expat-1.95.7.tgz' tar: lib/libexpat.so.5: Cannot stat: No such file or directory tar: Error exit delayed from previous errors pkg_create: make_dist: tar command failed with code 512 The missing so files are consistent with what I see when I try to build against an upgraded expat installation. The packaged version works just file: the so files are installed and all is well. But portupgrade wants to upgrade expat, and when it does, those missing files make everything else fail. I have no idea how to resolve the issue at the port level: is there a workaround? and I have been pulling from CVS so this shouldn't be an issue of being out of sync, I don't think. ruby is at 1.8.1, per someone else's advice, as well. -- Paul Beard paulbeard [at] mac.com
wrestling with b*rked package/ports collection
I seem to be in a loop where expat and gettext will somehow not install in such a way as to serve as valid dependencies. I have rebuilt either directly or as part of building something else countless times, as well as installing from a package. For whatever reason, expat seems to be missing something: Error: shared library "expat.5" does not exist portupgrade -f, installing from source, installing from a package: all seem to fail. And since it and/or gettext are core dependencies for everything else, it seems, there's not much progress being made. Any other sage advice? -- Paul Beard paulbeard [at] mac.com
Re: port questions (why do I find myself fudging symlinks to make stuff work?)
On Mar 30, 2004, at 12:23 AM, Matthew Seaman wrote: A more interesting question would be what output do you get from: % pkg_info -g jpeg-\* tiff-\* [/usr/local/lib]# ls -l libjpeg* libtiff* -rw-r--r-- 1 root wheel 159384 Mar 30 08:04 libjpeg.a -r--r--r-- 1 root wheel 354610 Mar 29 21:11 libtiff.a [/usr/local/lib]# pkg_info -g jpeg-\* tiff-\* Information for jpeg-6b_2: Mismatched Checksums: pkg_info: /usr/local/lib/libjpeg.so doesn't exist pkg_info: /usr/local/lib/libjpeg.so.9 doesn't exist Information for tiff-3.6.1_1: Mismatched Checksums: pkg_info: /usr/local/lib/libtiff.so doesn't exist pkg_info: /usr/local/lib/libtiff.so.4 doesn't exist Hmm, portinstall -f on those two ports seems to get caught in some recursive loop. I'm going thru my entire installed base and redoing it all. Perhaps there's a more clueful way to do it, but unless this will do any harm, it seems the most thorough. I need to explore the docs and get a better understanding of the ports tools: I seem to find this happening again and again. Thanks for the quick and detailed reply. -- Paul Beard paulbeard [at] mac.com
port questions (why do I find myself fudging symlinks to make stuff work?)
[ please cc me as I am not on the list] I have been trying to upgrade cups to 1.1.20 and have been running into some problems with libraries not being found (specifically jpeg.9 and tiff.4). After a few different iterations of pkg_add, portupgrade, portinstall, and accompanying pkg_delete, pkg_deinstall, make install, etc., I finally tried symlinking the .so files to the numbered versions that don't seem to exist and all seems to be well. ln -s /usr/local/lib/libjpeg.a /usr/local/lib/libjpeg.so.9 ln -s /usr/local/lib/libtiff.a /usr/local/lib/libtiff.so.4 pkgdb doesn't seem to have any problems (I have run it enough times today), cvsup has done its magic a couple of times. I thought portinstall/upgrade -rR would fix any out-of-date ports and make everything happy? What am I doing wrong to make this libraries not get installed? -- Paul Beard paulbeard [at] mac.com
Re: MySQL 3 server problems: mysqlclient.10 not found?
On Mar 16, 2004, at 11:00 PM, Ryan Merrick wrote: What is in #/usr/local/lib/mysql ? You should have something like: ... lrwxrwxrwx 1 root wheel 20 Feb 28 22:29 libmysqlclient.so -> libmysqlclient.so.10 -rwxr-xr-x 1 root wheel 132515 Nov 7 17:57 libmysqlclient.so.10 [/usr/home/paul]:: ls -l /usr/local/lib/mysql | grep mysql lrwxr-xr-x 1 root wheel 37 Mar 16 11:04 libmysqlclient.10 -> /usr/local/lib/mysql/libmysqlclient.a -rw-r--r-- 1 root wheel 188848 Oct 5 19:24 libmysqlclient.a lrwxrwxrwx 1 root wheel 20 Mar 16 16:06 libmysqlclient.so -> libmysqlclient.so.10 -rwxr-xr-x 1 root wheel 132515 Oct 5 19:24 libmysqlclient.so.10 -rw-r--r-- 1 root wheel 222510 Mar 16 14:08 libmysqlclient_r.a I ended up using pkg_add to fix this: not entirely satisfactory, but I'm up and running. -- Paul Beard paulbeard [at] mac.com
MySQL 3 server problems: mysqlclient.10 not found?
I have been trying to complete a portupgrade and something seems to be wrong: mysql323-server won't build. It seems to have a dependency on a file -- mysqlclient.10 -- that's part of mysql323-client but isn't found. ===> Compressing manual pages for mysql-client-3.23.58_2 ===> Running ldconfig /sbin/ldconfig -m /usr/local/lib/mysql ===> Registering installation for mysql-client-3.23.58_2 ===> Returning to build of mysql-server-3.23.58_2 Error: shared library "mysqlclient.10" does not exist *** Error code 1 Stop in /opt/ports/databases/mysql323-server. -- Paul Beard paulbeard [at] mac.com
Re: Yahoo! and GAIM
On Jan 9, 2004, at 8:27 PM, Eric F Crist wrote: I have those packages installed, just MUCH newer versions. How do I work around this and force the install? any reason not to build from the port? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Samba, Cups and Printing: Windows client gives "Access denied"
On Jan 1, 2004, at 9:33 AM, Peter Ulrich Kruppa wrote: I have got the same problem, only the other way round: Last week I managed to get my standalone print server with samba and cups working, but I can't exactly tell how I did it, so you can help me there :-) This is my smb.conf Hey, thanks. I was having this same problem and your smb.conf helped me out. I added these two lines to the stanza for the printer itself: guest ok = yes use client driver = yes And that seemed to make my "access denied" error go away. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Plan 9 style backup utility for FreeBSD?
On Jan 2, 2004, at 9:26 AM, Eric Rescorla wrote: I'm finally getting to the point where my disk capacity massively outruns my tape capacity, so I'm thinking of converting to removable disk-only backup. I could just use Amanda to backup to disk, but I'm intrigued by Plan 9's archival filesystem where backups from (say Jan 1, 1999) would go in /1999/1/1 and you could just find the files directly rather than grovelling through dump files. Is there some standard tool for doing this on FreeBSD? You can't just do a direct copy cause this results in major disk space wastage so I'm imagining you need to use hard links or something to keep the size down. I followed this on your weblog and I may be missing something: I keep coming back to some cocktail of find(1), ctime(3), diff(1) and friends (pax, tar) to locate and mirror only changed files in a duplicated hierarchy. If I'm understanding it, in 1999/1/1 you might have a file tree that looks like the "live" one with only files that were modified on 1999/1/1, but all other files would be links back to their unchanged versions. I like the idea of it. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Automated reply from mattm@mail.citystamp.com
On Dec 30, 2003, at 11:20 AM, Eric F Crist wrote: Is there a way we can filter out these types of messages or perhaps can someone tell me how to do it fairly easily? I think your better autoresponders don't send these to mailing lists: not sure why these are getting through. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: 5.2 where is /kernel.GENERIC
On Dec 29, 2003, at 3:08 PM, fbsd_user wrote: You are talking about the kernel source. I am talking about the /kernel.GENERIC binary execution file. locate kernel.GENERIC and see what it turns up. find / -name kernel.GENERIC -print cd / && ls -l kernel* -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Does FreeBSD allow one to use the Floppy Drive on a SunBlade 100 ?
On Dec 29, 2003, at 3:05 PM, fbsd_user wrote: I have Googled and I know that FreeBSD 4.9 cannot utilize the floppy drive as well. You are complete wrong. I have been using FBSD since 4.2 and the floppy has been working for me. I think the OP means on his specific hardware, the SunBlade 100. No one would dispute using a floppy on x86 hardware. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: How can I automatically mount a windows-xp share from a freeBSD box using windows username andpassword?
On Dec 29, 2003, at 10:58 AM, Joachim Dagerot wrote: I know this question has been touched earlier this year, but no answer came up at that point. So basically, has anyone been able to mount a windows share from a script in freeBSD? If yes, how did you do? I see this has been answered already, but there is a wealth of good information here: http://www.google.com/search?q=nsmbrc&ie=UTF-8&oe=UTF-8 -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: How can I automatically mount a windows-xp share from a freeBSD box using windows username andpassword?
On Dec 29, 2003, at 10:58 AM, Joachim Dagerot wrote: I know this question has been touched earlier this year, but no answer came up at that point. So basically, has anyone been able to mount a windows share from a script in freeBSD? If yes, how did you do? I have done it in Mac OS X so the procedure should be similar: you need to use .nmbrc to store your authentication details, but it should "just work." man nsmbrc for more specifics. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
kern.maxfiles problem resolved
I figured this out last night but not before I had one spell of catatonia on my system. I used mrtg to graph the increase in file descriptor usage and it made nice straight line, not a curve. That looked suspicious. I also set up a simple while loop in my shell to keep an eye on this. (while [ 1 ] ; do sysctl kern.maxfiles; sleep 60; done ). When that stopped updating at about 6600 files (the max is set at 16384), I rebooted (console was dead). In the process of looking at this, I had noticed that there were a few nmbd (samba) processes in the process table: I expect to see one or two, but not 10 or more. I watched this after rebooting, and sure enough, new processes were spawning ever couple of minutes, and this with no logins to the samba shares. What struck me as odd about this is that the abuse of the file table was being blamed on the wrong UID (80): nmbd doesn't run as the www user, so even if I had been more clueful about fstat, there's a good chance I would have been looking in the wrong places. I killed the samba processes, deinstalled samba, refreshed from cvs and, noting that the version was the same (2.2.8a), I reinstalled with portinstall -P. That seemed to do it. Now openfiles are sitting in the low 200s . . . . I've posted this on my weblog with the relevant image (didn't want to send an attachment to the list). http://www.paulbeard.org/movabletype/archives/001347.html -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: kern.maxfiles questions
On Dec 29, 2003, at 7:24 AM, Lowell Gilbert wrote: Is there some way to find out what's tying up the file descriptors? fstat(1) Of course. I had been thinking of that in terms of files, not file handles. D'oh. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: kern.maxfiles questions
Some updates to this. The machine was unresponsive this AM so I had to reboot it. I'm used to 90-100 day uptimes (as the power company permits) so this is quite unusual. On Dec 27, 2003, at 11:03 PM, paul beard wrote: [/usr/home/paul]:: sysctl kern.openfiles kern.maxfiles kern.openfiles: 257 kern.maxfiles: 4040 I raised the kern.maxfiles to 16383 and am monitoring it with mrtg (http://www.paulbeard.org/mrtg/red/red-openfiles.html). Already it's at almost 3000 (from less than 300 when I first noted it), so I assume I'll hit 4040 before too long. and on further examination I am finding some swap shortages (this machine has 512 Mb of real memory but only the same amount of swap). I actually only have 256 Mb of RAM with a half gig swap partition. I added another 512 Mb swap file and I am monitoring swap vs real mem usage as well (http://www.paulbeard.org/mrtg/red/red-mem.html). if this is all I need to do, I guess that's OK, but I'm still not sure why it became an issue all of a sudden. Is there some way to find out what's tying up the file descriptors? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
kern.maxfiles questions
I'm still having some issues with file descriptors being used up. Symptomatically, my http listener will become sluggish and some cgi processes I use will time out. The logs have a raft of these messages: Dec 27 22:46:13 red /kernel: kern.maxfiles limit exceeded by uid 80, please see tuning(7). Dec 27 22:47:12 red /kernel: kern.maxfiles limit exceeded by uid 1004, please see tuning(7). I can't even query sysctl, but after stopping httpd and postfix (the processes being run by those UIDs) I can finally get this information. [/usr/home/paul]:: sysctl kern.openfiles kern.maxfiles kern.openfiles: 257 kern.maxfiles: 4040 and on further examination I am finding some swap shortages (this machine has 512 Mb of real memory but only the same amount of swap). what tunables can be twiddled, assuming swap doesn't solve this? FreeBSD red.paulbeard.org 4.9-RELEASE FreeBSD 4.9-RELEASE #3: Sun Dec 21 14:01:26 PST 2003 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/RED i386 -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ghostview missing libc.so.4?
On Dec 27, 2003, at 10:27 PM, Chip Wiegand wrote: Mine has libc.so.5. No wonder the app is failing, it specifically wants .4 and is too dumb to work with anything newer. I'm getting real frustrated with this again, I'd like to move away from MS but it's not easy to do when the apps I need don't even work properly. You could fake it out with a symlink, but the Real Solution may be to rebuild whatever that file is part of, which I assume means a buildworld. I'm only running 4.9: I assume you're on 5.x if you have a higher file number? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Dynamic DNS Updates
On Dec 26, 2003, at 5:59 PM, Evan Sayer wrote: I would like to know how to make my server running 4.9 update my networks Dynamic Ip address on my domains DNS servers. My ip is always changing because it's dynamic, so a domain is essentially pointless unless i can get this up and running. You could say i want my domain to function as though it were assigned to a Static IP. I have been googleing for hours, and i can't understand any of it. Please be advised, i am somewhat new to FreeBSD. If you have a domain, I think you need to look into something like zoneedit.com: the dyndns and no-ip,org solutions, at least for free, don't cover domains outside their own namespace. With zoneedit you can have up to 5 domains for free. To keep them in sync with your dynamic host, you could use zoneclient.py (http://zoneclient.sourceforge.net/). It's worked fine for me so far (touch simulated woodgrain). -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ghostview missing libc.so.4?
On Dec 27, 2003, at 8:17 PM, chip wrote: I just installed ghostview and when I try to open a .ps file I get the error that libc.so.4 is missing. If this is a dependency then why wasn't it installed with ghostscript or ghostview? What do I need to do to fix this? There is no one port for libc.so.4. This is what I have for libc (which I think is pretty darn integral to your system): what does ls -l /usr/lib/libc.so* show? lrwxr-xr-x 1 root wheel 9 Nov 22 23:04 /usr/lib/libc.so -> libc.so.4 -r--r--r-- 1 root wheel 578964 Nov 22 23:04 /usr/lib/libc.so.4 -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
kern.maxfiles limit exceeded: what to investigate?
I upgraded my webserver hardware from a old Pii 233 to an AMD Athlon 700 a few weeks back and seem to be having some teething troubles with it. I have hit the kern.maxfiles limits twice recently, having run for a couple of years without even knowing there was one. I found 420 of these in messages: Dec 21 13:39:30 red /kernel: kern.maxfiles limit exceeded by uid 80, please see tuning(7). and I could only login on console. I read thru tuning(7) and the best I could figure is that my best option was to let the system work its limits based on hardware: accordingly, I set maxusers in my config to 0. I didn't understand the rest of it and I'm not sure much of it applies, since it's not exactly loaded (less than 1 hits/day). This is what I see now. It looks like a lot of headroom, I think. kern.maxfiles: 4040 kern.maxfilesperproc: 3636 kern.openfiles: 300 Since UID 80 would be the httpd process, I suppose looking into process-specific resource issues is next. I am running Apache 1.3.29. One thing I noticed that seemed a little odd was the snmpd seemed to behaving strangely. I did an snmpwalk to see if I could monitor these kernel values that way (there is so much useful stuff exposed thru snmp), but I found that I couldn't run it more than once, and that snmpd was running at 97% or so of CPU. I have deinstalled and rebuilt it, and now it seems to be behaving properly, but I wonder if every 5 minute snmp requests, in and outbound, with a flaky binary were slowly eating up file descriptors. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Page faults every few days
On Dec 13, 2003, at 9:19 PM, Jaime wrote: (I don't see why a user-space process would interfere with kernel resource management.) me, either, but it seemed worth looking into. since this seems to be related to swapping, how much real memory vs swap do you have? do you have any idea how much swapping is going on? I'm sure some more knowledgeable people than supply better diagnostic questions, but what do these values look like on the box you're concerned with? vm.stats.vm.v_swapin: 36869 vm.stats.vm.v_swapout: 63458 vm.stats.vm.v_swappgsin: 119489 vm.stats.vm.v_swappgsout: 283967 -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Page faults every few days
On Dec 13, 2003, at 8:50 PM, Jaime wrote: current process = 26642 (perl5.00503) any idea what perl-based job is running and if moving to a newer perl version would have any effect at all? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Perl 5.8.2 problems (was Re: how to build Spamassassin)
On Dec 11, 2003, at 9:54 AM, Tony Jones wrote: I'm very unfamiliar with the ports system. I've never heard of portinstall or portupgrade. Just running make && make install in the appropriate port subdirectory. It seems to me you're making this really complicated: I don't know what difference it makes where things get installed (/usr/local/{language}/ . . . ), but I have been using the ports collection as it comes without a problem. One of the benefits of using a system (like FreeBSD) is that there are some design conventions and decisions you can rely on. My advice, and worth every penny you're paying, would be to use the ports system and become familiar with it before hacking around it. To that end, I would do, as root" cd /usr/ports/sysutils/portupgrade; make install and then use that to manage the rest of it with portinstall , in your case perl5.8 and spamassassin. You may need to run "use.perl port" between those steps to ensure that spamassassin gets built against perl5.8 and doesn't complain about the wrong version. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Perl 5.8.2 problems (was Re: how to build Spamassassin)
On Dec 10, 2003, at 8:34 PM, Tony Jones wrote: At this point, /usr/local/perl/bin/perl is installed why is it in /usr/local/perl/bin? As far as I have seen, the ports collection doesn't do that. did you install as a port (make install in /usr/ports/lang/perl5.8)? What would happen if you were to use portinstall perl5.8 and portinstall spamassassin? You may need to install the portupgrade package if you haven't already done so. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
how to make dhclient give up its lease and get a new one with routes, etc. ?
I am on a cable modem and am running an http server, natd, the Usual Suspects, and have been doing this for a couple of years. I use dhcp to get the address for my externally-facing NIC and through reboots, power outages, even changes in network ownership (@ home -> ATT -> comcast) all has been solid. Until today. I decided over the past couple of weeks to actually get a domain instead of using no-ip.org (paulbeard.org: how imaginative) and I started using Zoneedit to manage it since I wanted DNS to be somewhere other than on my network. As luck would have, I got assigned a new IP address last night (for the first time since I used @home more than 3 years ago) and for some reason the dhcp client wouldn't get the new address. I ran dhclient and that just set my address to all 0's. I then killed it and restarted and got an address but no routes. I ended up rebooting and that solved it (I then had to fix both no-ip's config file since the old one was unparseable by version 2 clients and go to zoneedit and update my address). So how to make dhclient do all this without the Windows workaround of rebooting? Advice on automagically notifying zoneedit is welcome, as well but not as pressing: I think they work with dyndns and of course I chose no-ip.org. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: NATd question
On Dec 7, 2003, at 11:15 PM, Matt Edwards wrote: consumer: "I have two computers. I need to make sure they can both get on the internet." (Thinking: "I know my buddy did this with his setup") ISP: "Oh you mean you need a second IP address, right?" (Thinking: "The poor guy doesn't know he can do it with one and NAT server. But I ain't telling him that.") Of course, not long ago, you would have had the 1st tier tech support drone accuse you of stealing bandwidth if you mentioned that you had more than one machine networked. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Why would drive run at UDMA33? (Segate 80GB)
On Dec 7, 2003, at 10:07 PM, JacobRhoden wrote: I typed 'man atacontrol' and it didnt seem to help! (grin). Seriously, i had a look and worked out how to list/display the modes of drives, and i am not sure what commands i would type to help fix the speed? mode Without the two mode arguments, the current transfer modes of both devices are printed. If the mode arguments are given, the ATA driver is asked to change the transfer modes to those given. The ATA driver will reject modes that are not supported by the hardware. Modes are given like ``PIO3'', ``udma2'', ``udma100'', case does not matter. If one of the devices mode should not be changed, use a nonexisting mode as argument (i.e. ``XXX''), and the mode will remain unchanged. Currently supported modes are: BIOSDMA, PIO0 (alias BIOSPIO), PIO1, PIO2, PIO3, PIO4, WDMA2, UDMA2 (alias UDMA33), UDMA4 (alias UDMA66), UDMA5 (alias UDMA100) and UDMA6 (alias UDMA133). So you can type 'atacontrol mode ' and change the speed, at some potential risk, as the man page warns. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: the proces that can be monitored
On Dec 5, 2003, at 7:03 AM, M.D. DeWar wrote: I'm at the point were it asks for process to monitor but am not sure what ones I can and what they are are called. In the example says httpd for http so I know that one but not sure what it is for memory disk space It might be worth running snmpwalk against the host you want to monitor and see what's available. There are a lot of variables to look at. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: config(8) out of sync with source
On Dec 3, 2003, at 7:58 PM, William O'Higgins wrote: I hadn't used that method, because I was unsure of the results, and the procedure in the Handbook is more extensive than I had time for. My reticence was well founded, since when I followed the above procedure it rendered the computer non-bootable. I doubt it's the procedure that had that effect: I've built kernels and worlds on several machines (desktops and laptop) since 4.2 RELEASE and I think I had one dud kernel in all the time. I'm not sure how to interpret the comment that "the procedure in the Handbook is more extensive than I had time for." It doesn't take any time, really, unless you sit and watch the compiler messages streaming past. If you haven't tried this yet, I recommend following the Handbook method with a GENERIC kernel and see how that goes. That should give you a kernel remarkably similar to the one you installed initially. If that works, you can rest assured the procedure is sound, and it's the configuration that needs tweaking. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: ports problems
On Dec 1, 2003, at 7:28 PM, RYAN vAN GINNEKEN wrote: when i run portsdb -Uu i get many errors like the one below is there anyone who can help me try pkgdb -F and see how that goes. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: games that don't need X-11
On Dec 1, 2003, at 1:28 PM, Marty Landman wrote: I'm interested in porting some games, but want those that do not involve heavy graphics dependencies. Any suggestions or urls that might list these, or hints on how I can examine the dependencies on my own? Do you mean installing existing games in the ports tree or porting other games that are not there yet? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Remove ^M characters from xhtml file
On Nov 30, 2003, at 6:53 PM, Melvyn Sopacua wrote: Another way: perl -pi.bak -e 's/\r$//' *.xhtml BTW: why is this even an issue that needs a solution? XHTML doesn't care one way or the other, since all linear spacing is folded into one space. It's distracting to look at all that stuff if you're editing in vi, for example. But for yet one more solution, you can search and replace ^M as Control V Control M in vi. I also use perl -pi "s#\\r#\\n#g" all the time. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: adaptive stealth in ipfw?
On Nov 28, 2003, at 10:18 AM, W. D. wrote: I'd be very interested in seeing some fair criticism of what Steve Gibson is doing. However, www.GRCsucks.com seems to have a number of broken links. Where the links work, the verbiage seems to be more confusing than clarifying. I found that to be the case as well. It looks like there is a kernel of truth to some of the allegations (ie, a false sense of security is worse than no security at all), but if anyone can recommend a reliable and accurate security scanner (other than a friend with netsaint), could they share it? Thanks. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Adding new IP's without reboot?
On Nov 26, 2003, at 4:32 PM, Ben Dover wrote: Is there a way to add new IPs to a FreeBSD 4.9 or 5.1 box without rebooting. I add them to /etc/rc.conf but they are not effective until a reboot. There are some webhosting assistant programs which allow instant use of IPs with *nix and I was hoping there was a way to do this in FBSD. This question was the basis for a useful thread a week ago: googling or sifting through the archives should turn up some useful information. I think every variant of UNIX supports this with ifconfig: something like this (depending on your interface device name). [/usr/home/paul]:: ifconfig xl1 192.168.2.100 up -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: POLA violation?: snmp renumbering stuff
On Nov 23, 2003, at 6:00 PM, Dan Nelson wrote: I don't think snmp tables have any defined order. I don't even know if the index for a particular resource is guaranteed to be stable across filesystem dismount/remounts. Something like this should work: My issue was that they shouldn't change once defined: otherwise, how can you reliably use something if it adopts different behavior with each new release/build? After all, we're not talking about Windows here . . . . ;-) It would be useful if / were always 1, for example. It looks like, with the inclusion of RAM and swap in the table, / might be 3. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: remote mount hangs sysstem
On Nov 23, 2003, at 1:31 PM, RYAN vAN GINNEKEN wrote: So if i set to noauto does it still get mounted or do i have to execute a command later sure, just use "mount /mount/point" If the fstab works now, you won't need to do anything else. There may be more sophisticated ways (automounters and such that mount the filesystem as you traverse into it) but I've never used them. man 8 amd, for example. DESCRIPTION Amd is a daemon that automatically mounts filesystems whenever a file or directory within that filesystem is accessed. Filesystems are automati- cally unmounted when they appear to be quiescent. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: remote mount hangs sysstem
On Nov 23, 2003, at 12:13 PM, RYAN vAN GINNEKEN wrote: Is there a way to mount a cdrom or remote file systems using fstab but not having it crash out the system. example if i have a nfs share set up to another machine and that machine goes down the next time i reboot my system the machine hangs when it cannot find the share and will not allow me to do anything and i have to hook up a monitor and keyboard to get it back the same happens when there is an error on a cd rom make sure the entry on /etc/fstab is set to noauto: then the client won't try to mount the filesystem on boot. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
POLA violation?: snmp renumbering stuff
For some reason, my locally installed snmp daemons decided to renumber the elements in the hrStorageTable, meaning all the attached disks were being either misreported or just plain dropped from my graphs (paulbeard.no-ip.org/mrtg/blue/index.html). Not that the new numbering doesn't make sense but I didn't know this was going to happen. How to discover and fix it? snmptable is my friend. As shown here, the memory used by the kernel is listed first, followed by the disks. The disks were numbered starting at 1 before . . . . . [/www/mrtg/blue]# snmptable -c blue hrStorageTable SNMP table: HOST-RESOURCES-MIB::hrStorageTable hrStorageIndex hrStorageType hrStorageDescr hrStorageAllocationUnits hrStorageSize hrStorageUsed hrStorageAllocationFailures 1 HOST-RESOURCES-TYPES::hrStorageOther Memory Buffers 256 Bytes ? 192 0 2 HOST-RESOURCES-TYPES::hrStorageRam Real Memory 4096 Bytes ? 3241 ? 3 HOST-RESOURCES-TYPES::hrStorageVirtualMemory Swap Space 4096 Bytes ? 19625 ? 4 HOST-RESOURCES-TYPES::hrStorageFixedDisk / 1024 Bytes ? 83592 ? 5 HOST-RESOURCES-TYPES::hrStorageFixedDisk /usr 1024 Bytes ? 3639961 ? 6 HOST-RESOURCES-TYPES::hrStorageFixedDisk /var 1024 Bytes ? 8015 ? 7 HOST-RESOURCES-TYPES::hrStorageFixedDisk /proc 4096 Bytes ? 1 ? 8 HOST-RESOURCES-TYPES::hrStorageFixedDisk /usr/ports 512 Bytes ? 35548516 ? -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: advice needed on creating hmtl docs?
On Nov 23, 2003, at 9:22 AM, fbsd_user wrote: I would really like to use the ms/word docs as source input to some FBSD hmtl generator to build original hmtl source that apache can serve up natively. Can any body suggest how to do this? One approach I've had some success with is to take the baroque HTML that Word generates and run it through htmltidy. It can strip out all the deprecated tags and generate CSS styles for you, giving you both the appearance and the maintainability you may need later. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: make vs. pkg_add
On Nov 22, 2003, at 8:00 PM, Patrick Burnett wrote: I'm probably doing something wrong such that 'make' isn't playing nice, but I'd still appreciate some further insight from more experienced users. Perhaps you could explain more about what difficulties you're having. I have used both pkg_add and make install clean interchangeably, though I prefer working from source. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Perl configuration
On Nov 22, 2003, at 3:11 AM, Matthew Seaman wrote: However, the values in make.conf only have an effect at compile time, so if you want all of your perl modules to live under /usr/local/lib/perl5/{,site_perl/}5.8.2 then you're going to have to reinstall all of the ports that put files into %%SITE_PERL%% as well as installing the updated version of perl. Aargh, I had this happen to me on two boxes in the past week so the scars are still fresh enough for me to remember it all. What worked for me [tm] was to invoke portinstall -r perl58 and let it fix everything. That worked when a simple "make [re]install" wouldn't. (the r argument is valuable: -r, --recursive Do with all those depending on the given packages as well ) My advice, worth every penny you pay for it, is to leave the extra directories alone (disk space is cheap). The side-effects are too hard to predict or rectify, in my experience. -- Paul Beard paulbeard [at] mac.com ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: FreeBSD Motherboard survey...
On Nov 21, 2003, at 9:53 AM, Philip Hallstrom wrote: Hi all - I'm wanting to build my own computer to run FreeBSD, but don't have the slightest idea (well, maybe the slightest :) what motherboard to buy. I'd like one that has built-in lan/audio that works in FreeBSD, but in my searching efforts I've turned up very little. I know the handbook lists chip sets and whatnot, but trying to figure out what chip set is on a given board isn't always easy -- at least not for me. So... I wrote a survey app that I'm hoping lots of you will fill out. It let's you pick the brand/model of motherboard, then indicate what onboard features work (lan, audio, video, usb, firewire, ide, sata, scsi, raid, smp), lets you provide some overall "satisfaction" ratings, and then provide any additional comments. As a side effect you can input your laptop information which seems to be a frequently asked source of questions. If I get enough responses I'd like to expand this to include other things such as DVD players, USB peripherals, etc. that have varying levels of support in FreeBSD. Perhaps turn it into a companion to the HARDWARE.TXT document... Anyway.. here it is: http://www.eilio.com/freebsd-motherboards/ This is very nice, and I think it will be helpful to a lot of people who want to move to FreeBSD. I especially like the idea of laptop users being able to contribute. This is a topic that has come up many times on the mobile@ list and I think a structured form where components can be detailed (works/broken) is a great idea. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Can I install packages only for my release?
On Nov 20, 2003, at 5:06 PM, Rahul Fernandez wrote: Hi, I'm am running 4.9 release. A package called hpijs1.4.1 is installed. I now would like to upgrade to hpijs-1.5. However, this package is only available in 4.9-stable. Can I install the package from 4.9-stable or is it advisable to stick to the packages in my release? why not install it from a port? I use packages as a last resort, and go with ports first. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Network messaging
On Nov 19, 2003, at 1:21 PM, Feltis, Ralph C. wrote: Ahh yes, I see how preparing an elegant solution would provide greater flexibility. However, Cordula's solution (ssh [EMAIL PROTECTED] strings somefile) is about a 15 second fix, whereas setting up a server is, well, not so much a 15 second fix. Thanks for your all help guys. This is something you could find in the Perl Cookbook (or even in the man pages, if I remember rightly). Or you could do it really crudely and have a shell script that consists of "cat /some/text/file" set up in /etc/services and [x]inetd to spit out the information you want. Depends on how secure it needs to be or how involved you want to get. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Network messaging
On Nov 19, 2003, at 11:51 AM, Feltis, Ralph C. wrote: Is it possible to set some type of simple messaging system between networked FreeBSD machines? For example, client A pings client B, and then client B sends some prespecified text reply to client A. to what end? I don't know how you respond to a literal ping, but you could code up a simple client <-> server system where client A connects to client B on a specified port and B emits some text on request. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Monitoring
On Nov 19, 2003, at 8:59 AM, Rus Foster wrote: Does anyone have a recommendation for a good system monitoring (swap,I/O monitor) from ports? Ideally I want something that runs out of cron and emails me if usage goes above a certain point net/mrtg has some threshold/alerting stuff builtin in addition to it's graphing/trend watching capabilities. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: freebsd-questions@freebsd.org
On Nov 16, 2003, at 2:40 AM, [EMAIL PROTECTED] wrote: I has a G3 iMac that has broken CD-Rom drive, can it be reformated comppletely from the web? To expand on another answer, you can install NetBSD for PowerPC over the net, but I think you need a floppy or some kind of startup image to boot from. And you have to be willing to poke around in the Open Firmware environment. And of course, there's Darwin 7.0.1, the core of OS X 10.3. The best option is to fix or replace the drive, I think. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: FreeBSD Essay.
Lewis Thompson wrote: Hey guys, I'm a first year CS student at Manchester and I've been given the task of writing a 1,000 word essay on something computer-related. It can be pretty much anything I want (I think). I've decided FreeBSD is interesting, the OS I advocate and that I shall write about this. I am planning to write a brief history of the four BSDs, going way back in time (probably a few words on Ritchie, etc.) but then concentrate on FreeBSD. I think showing the pedigree is a good idea. I would extend it to discuss and demonstrate how FreeBSD is a system with solid design principles and a resourceful community around it. It's benefits are more than technical, as anyone on this list can attest. Given the short length, I think an overview of how FreeBSD came to be, it's strengths, and whatever seems appropriate in the projected evolution would be worth reading. I'm really asking if anybody can suggest any particularly interesting topics that I can go away and research and then include in my essay. I guess since it's only a short essay I can't have /too/ much detail and I didn't particularly want to try and explain something /very/ complicated (although please suggest just the same ;). -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 Who messed with my anti-paranoia shot? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Recommendations for wireless networking and FreeBSD
Toni Schmidbauer wrote: On Mon, Nov 03, 2003 at 10:14:19AM -0800, John DeStefano wrote: I've just moved into an apartment in which drilling and running wires is taboo. Has anyone delved successfully into the realms of wireless networking their FreeBSD groups? My main server is running 4.8-STABLE, and I have a client machine running 5.1-RELEASE (which has been suspect to a lack of driver support for its onboard NIC in FBSD anyway), but I am not married to any of these releases and would up/downgrade if a solution was available. I'd also prefer a Wireless-G access point and adapter solution if possible, as opposed to the much slower B solutions available. man 4 wi. there you can find a list of support cards. man 4 an has the straight dope on the aironet driver: I have been using it for awhile with FreeBSD 4.4 - 4.8. -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 A formal parsing algorithm should not always be used. -- D. Gries ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Mozilla 1.5 errors?
Lowell Gilbert wrote: paul beard <[EMAIL PROTECTED]> writes: What's wrong here? I've updated ports today, if that's useful to know. Just built fine for me. -STABLE. Anything funny in your make.conf? My problem was that some ports were out of sync. -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 A bachelor is a selfish, undeserving guy who has cheated some woman out of a divorce. -- Don Quinn ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Mozilla 1.5 errors [resolved?]
Kris Kennaway wrote: On Sun, Nov 02, 2003 at 02:02:31PM -0800, paul beard wrote: I ended up tracking down what port owned Xos.h [XFree86-4-libraries] and rebuilding/installing it. At that point, Mozilla would build and install just fine. However, it wouldn't run: it need libintl.so.4 and I have libintl.so.5 installed. As you've discovered, your installed packages are inconsistent. At some point you updated the gettext port without rebuilding everything that depends on it, so packages that were linked against the old library (libintl.so.4) can no longer function because you removed it. The fix is to rebuild everything that depends on gettext by doing e.g. portupgrade -r gettext. Roger that. Thanks for the solution. -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 Alden's Laws: (1) Giving away baby clothes and furniture is the major cause of pregnancy. (2) Always be backlit. (3) Sit down whenever possible. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Mozilla 1.5 errors [resolved?]
I reported some problems with building Mozilla [1.5] from ports, specifically this one: In file included from cppsetup.c:28: def.h:29: X11/Xos.h: No such file or directory def.h:30: X11/Xfuncproto.h: No such file or directory gmake[2]: *** [host_cppsetup.o] Error 1 gmake[2]: Leaving directory `/opt/ports/www/mozilla/work/mozilla/config/mkdepend' gmake[1]: *** [export] Error 2 gmake[1]: Leaving directory `/opt/ports/www/mozilla/work/mozilla/config' gmake: *** [default] Error 2 *** Error code 2 I ended up tracking down what port owned Xos.h [XFree86-4-libraries] and rebuilding/installing it. At that point, Mozilla would build and install just fine. However, it wouldn't run: it need libintl.so.4 and I have libintl.so.5 installed. The simple workaround was to symlink the needed file from the one I had and that seems to be working. libintl.so.* is installed by gettext. I have no idea what the Real Solution for this is, but I leave these workaround steps in the archive in case anyone finds them useful . -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 Greener's Law: Never argue with a man who buys ink by the barrel. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Mozilla 1.5 errors?
What's wrong here? I've updated ports today, if that's useful to know. ===> Building for mozilla-1.5_1,2 /usr/local/bin/gmake -C config export gmake[1]: Entering directory `/opt/ports/www/mozilla/work/mozilla/config' gmake[2]: Entering directory `/opt/ports/www/mozilla/work/mozilla/config/mkdepend' cppsetup.c cc -o host_cppsetup.o -c -O -pipe -DXP_UNIX -O3 -DINCLUDEDIR=\"/usr/include\" -DOBJSUFFIX=\".o\" -DPREINCDIR=\"include\" -I../../dist/include/mkdepend -I../../dist/include -I/opt/ports/www/mozilla/work/mozilla/dist/include/nspr -I/usr/local/include -I/usr/local/include -I/opt/ports/www/mozilla/work/mozilla/dist/include/nspr cppsetup.c In file included from cppsetup.c:28: def.h:29: X11/Xos.h: No such file or directory def.h:30: X11/Xfuncproto.h: No such file or directory gmake[2]: *** [host_cppsetup.o] Error 1 gmake[2]: Leaving directory `/opt/ports/www/mozilla/work/mozilla/config/mkdepend' gmake[1]: *** [export] Error 2 gmake[1]: Leaving directory `/opt/ports/www/mozilla/work/mozilla/config' gmake: *** [default] Error 2 *** Error code 2 -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 Remember, even if you win the rat race -- you're still a rat. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Fwd: Re: kernel "make" error.
Reply-to: risto phario <[EMAIL PROTECTED]> From: risto phario <[EMAIL PROTECTED]> TO: paul beard <[EMAIL PROTECTED]> CC: Date: Thu Oct 30, 2003 11:38:05 AM PST Subject: Re: kernel "make" error. Ok, I have attached the error msg. Thanks. Risto - Original Message - From: paul beard <[EMAIL PROTECTED]> Date: Wed, 29 Oct 2003 21:21:26 -0800 To: risto phario <[EMAIL PROTECTED]> Subject: Re: kernel "make" error. > risto phario wrote: > > Hi, I was hoping you could help me out a bit. I am new to > > FreeBSD and was trying to compile a custom kernel. I ran config > > on it and everything went good. I did make depend, and > > everything went good. Then I did make and it gave me an error. > > I am new to FreeBSD so I have no idea what to do. I have > > attched a copy of my custom kernel "MYKERNEL". Any help with > > this problem would be great!!! Thank you. > > Sending the error message would be a big help. > > > > -- > Paul Beard > <http://paulbeard.no-ip.org/movabletype/> > whois -h whois.networksolutions.com ha=pb202 > > Kansas state law requires pedestrians crossing the highways at > night to > wear tail lights. > -- ___ Get your free email from http://mymail.bsdmail.com Powered by Outblaze makeerr Description: Binary data ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: kernel "make" error.
risto phario wrote: Hi, I was hoping you could help me out a bit. I am new to FreeBSD and was trying to compile a custom kernel. I ran config on it and everything went good. I did make depend, and everything went good. Then I did make and it gave me an error. I am new to FreeBSD so I have no idea what to do. I have attched a copy of my custom kernel "MYKERNEL". Any help with this problem would be great!!! Thank you. Sending the error message would be a big help. -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 Kansas state law requires pedestrians crossing the highways at night to wear tail lights. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Should I upgrade to 4.9?
Dragoncrest wrote: The mail server is my only mission critical machine right now, so that's why I'm asking this. It's the only box I can't afford to have go down. I think you answer your own question. As a practical matter, I think 4.9 is a maintenance release on the 4.x branch, so if it ain't broke, don't fix it. -- Paul Beard <http://paulbeard.no-ip.org/movabletype/> whois -h whois.networksolutions.com ha=pb202 Reader, suppose you were an idiot. And suppose you were a member of Congress. But I repeat myself. -- Mark Twain ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "[EMAIL PROTECTED]"