Relayd performance issues
I am having some strange performance issues with a relayd.
I have a setup where relayd forward http picture requests to two webservers.
If you visit a webpage with the pictures handled by relayd (normally 6
pics), maybe 4 of these will load at once, and the last 2 will load a little
slow 1-2 secs.
But this is only an issue the first time you hit a webpage, if you goto the
next page with 6 new pictures they will all load at once.
If you then wait for 5 minutes and try again, you will see the same pattern
again, delays on the first page and then no delays on the next pages.
There is no problems when I hit the webservers directly without relayd.
Adding/removing the tcp performance options below from the config file
does not change anything
My config file:
interval 10
timeout 200
prefork 25
log updates
http protocol http_setup {
header append $REMOTE_ADDR to X-Forwarded-For
tcp { nodelay, sack, socket buffer 65536, backlog 100 }
}
table pixXX.xx.com { xxx.173.226.135, xxx.173.226.184 }
relay pixXX.xx.com {
listen on 213.173.226.142 port 80
forward to pixXX.xx.com port 80 mode loadbalance check tcp
}
Any ideas about what can cause this behavior?
Thanks,
Claus Larsen
Re: usr.sbin/wake removal
On 9/02/2009, at 6:31 PM, Thomas Pfaff wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. My guess would be that I want this 10K util, you want that 7.2K util, Fred wants that 20K util, and every Tom, Dick, and Harry wants their n K ... who gets to make the rules, who gets to administer it, etc.? (Who gets to listen to everyone arguing why this or that should go in?) And guess there may be ramifications for install media?
Re: usr.sbin/wake removal
On Mon, Feb 09, 2009 at 09:05:13PM +1300, Richard Toohey wrote: On 9/02/2009, at 6:31 PM, Thomas Pfaff wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. My guess would be that I want this 10K util, you want that 7.2K util, Fred wants that 20K util, and every Tom, Dick, and Harry wants their n K ... who gets to make the rules, who gets to administer it, etc.? (Who gets to listen to everyone arguing why this or that should go in?) And guess there may be ramifications for install media? you need wake(8) on install media to wake up your local ftp mirror? -- Alexander Yurchenko
Re: Is it possible to increase wscale multiplier?
On Sun, Feb 08, 2009 at 09:58:23PM +, Dieter wrote: Grepping through a few log files, the userland program read 44,751,896 bytes with a single syscall. The default recv buf size of 65536 doesn't get the job done for this application. OK, I'll take the bite. The following scenario might apply to your userland app: If your userland program has a very big buffer, it probably takes a while to process it once the read(2) call succeeds. In the meantime new data arrives (which might be dropped on the floor if the sending side does not back off quickly enough because of lack of acks): the kernel buffer fills up to the maximum receive size, and the userland buffer is not available because the new read call has not been issued yet. Not until now you have just given us vague assumptions and jump to conclusion statements without almost no facts about your app, so of course we jump the gun and bash you. If you expect us to help to solving a problem provide details, and not conclusions. You have no clue if my userland app is well behaved or not. And there IS a problem if another driver locks the Ethernet driver out for too long. I hunted down and beat into submission an example of that problem recently. See above. Until you give us details these statements have not useable content. Our socket buffers will never allow that amount of memory to be queued. I think Claudio doesn't know that Step 1 in solving userland throughput problems is to blame it on the kernel, hardware, drivers or actually anything except the application? Learn to read. It is a latency problem, not a throughput problem. See above. And I see the alternative all my problems would be solved if OpenBSD had feature X (in this case real-time support) is also used, so extra bonus points! Learn to read. I haven't tried it on OpenBSD yet, just FreeBSD and NetBSD. See above. Geez, ask an innocent question and suddenly get accused of not understanding anything. By people that didn't bother to read and clearly don't understand the problem. If anything, we do not understand the problem BECAUSE you are not giving details. -Otto
Re: bgpd fails to install ipv6 routes in kernel routing table
On Mon, Feb 09, 2009 at 04:51:12PM +1100, Graeme Lee wrote:
Graeme Lee wrote:
Graeme Lee wrote:
tico wrote:
Graeme Lee wrote:
tico wrote:
Graeme Lee wrote:
snip
Ok forget bgp configs for a minute. I've been quickly scanning over
the code, and notable is that the log displays:
Feb 9 13:00:15 gw-nextgen bgpd[17223]: send_rtmsg: action 1, prefix
2001:7fb:fe07::/48: Network is unreachable
but shouldn't it be a send_rt6msg call in kroute.c?
Yes. The waning message had the wrong function name in it.
On a hunch, I tried a 64bit and a 32 bit machine with 1 prefix each.
The 32bit machine adds routes to the kernel without complaint. The
64bit machine complained with send_rtmsg
Arrg. IPv6 is once again broken by design. For some ridiculous reason
struct sockaddr_in6's size is 28 bytes. So IPv6 fucks up alignment on 64 bit
archs. All hail link local addressing and all the crappy workarounds
needed for it.
Please try the attached diff.
--
:wq Claudio
Index: kroute.c
===
RCS file: /cvs/src/usr.sbin/bgpd/kroute.c,v
retrieving revision 1.164
diff -u -p -r1.164 kroute.c
--- kroute.c9 Feb 2009 08:20:11 - 1.164
+++ kroute.c9 Feb 2009 08:49:00 -
@@ -2057,12 +2057,14 @@ retry:
int
send_rt6msg(int fd, int action, struct kroute6 *kroute)
{
- struct ioveciov[5];
+ struct ioveciov[8];
struct rt_msghdrhdr;
struct sockaddr_in6 prefix;
struct sockaddr_in6 nexthop;
struct sockaddr_in6 mask;
struct sockaddr_rtlabel label;
+ chargrmbl[sizeof(long) - (sizeof(prefix)
+ (sizeof(long) - 1))];
int iovcnt = 0;
if (kr_state.fib_sync == 0)
@@ -2070,6 +2072,7 @@ send_rt6msg(int fd, int action, struct k
/* initialize header */
bzero(hdr, sizeof(hdr));
+ bzero(grmbl, sizeof(grmbl));
hdr.rtm_version = RTM_VERSION;
hdr.rtm_type = action;
hdr.rtm_tableid = kr_state.rtableid;
@@ -2096,6 +2099,11 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = prefix;
iov[iovcnt++].iov_len = sizeof(prefix);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
+
if (memcmp(kroute-nexthop, in6addr_any, sizeof(struct in6_addr))) {
bzero(nexthop, sizeof(nexthop));
@@ -2110,6 +2118,10 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = nexthop;
iov[iovcnt++].iov_len = sizeof(nexthop);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
}
bzero(mask, sizeof(mask));
@@ -2123,6 +2135,10 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = mask;
iov[iovcnt++].iov_len = sizeof(mask);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
if (kroute-labelid) {
bzero(label, sizeof(label));
Re: bgpd fails to install ipv6 routes in kernel routing table
Claudio Jeker wrote: On Mon, Feb 09, 2009 at 04:51:12PM +1100, Graeme Lee wrote: Graeme Lee wrote: Graeme Lee wrote: tico wrote: Graeme Lee wrote: tico wrote: Graeme Lee wrote: snip Ok forget bgp configs for a minute. I've been quickly scanning over the code, and notable is that the log displays: Feb 9 13:00:15 gw-nextgen bgpd[17223]: send_rtmsg: action 1, prefix 2001:7fb:fe07::/48: Network is unreachable but shouldn't it be a send_rt6msg call in kroute.c? Yes. The waning message had the wrong function name in it. well I was looking at least. On a hunch, I tried a 64bit and a 32 bit machine with 1 prefix each. The 32bit machine adds routes to the kernel without complaint. The 64bit machine complained with send_rtmsg Arrg. IPv6 is once again broken by design. For some ridiculous reason struct sockaddr_in6's size is 28 bytes. So IPv6 fucks up alignment on 64 bit archs. All hail link local addressing and all the crappy workarounds needed for it. Please try the attached diff. You are altogether a legend. I now have the full ipv6 table in the kernel.
Re: bgpd fails to install ipv6 routes in kernel routing table
Am 09.02.2009 um 09:53 schrieb Claudio Jeker: Please try the attached diff. A general question about diffs like this: will these diffs automatically go to -current in the next couple of days/weeks? Or do I have to apply all these patches by hand? :wq Claudio Thanks, Falk
Re: bgpd fails to install ipv6 routes in kernel routing table
On Mon, Feb 9, 2009 at 12:53 AM, Claudio Jeker cje...@diehard.n-r-g.com wrote:
On a hunch, I tried a 64bit and a 32 bit machine with 1 prefix each.
The 32bit machine adds routes to the kernel without complaint. The
64bit machine complained with send_rtmsg
Arrg. IPv6 is once again broken by design. For some ridiculous reason
struct sockaddr_in6's size is 28 bytes. So IPv6 fucks up alignment on 64 bit
archs. All hail link local addressing and all the crappy workarounds
needed for it.
Maybe it is too late for me to be thinking about this ... but could
you explain the diff below? Unless I'm missing something obvious, it
looks like it changes behavior for non-64bit archs as well.
--patrick
Please try the attached diff.
--
:wq Claudio
Index: kroute.c
===
RCS file: /cvs/src/usr.sbin/bgpd/kroute.c,v
retrieving revision 1.164
diff -u -p -r1.164 kroute.c
--- kroute.c9 Feb 2009 08:20:11 - 1.164
+++ kroute.c9 Feb 2009 08:49:00 -
@@ -2057,12 +2057,14 @@ retry:
int
send_rt6msg(int fd, int action, struct kroute6 *kroute)
{
- struct ioveciov[5];
+ struct ioveciov[8];
struct rt_msghdrhdr;
struct sockaddr_in6 prefix;
struct sockaddr_in6 nexthop;
struct sockaddr_in6 mask;
struct sockaddr_rtlabel label;
+ chargrmbl[sizeof(long) - (sizeof(prefix)
+ (sizeof(long) - 1))];
int iovcnt = 0;
if (kr_state.fib_sync == 0)
@@ -2070,6 +2072,7 @@ send_rt6msg(int fd, int action, struct k
/* initialize header */
bzero(hdr, sizeof(hdr));
+ bzero(grmbl, sizeof(grmbl));
hdr.rtm_version = RTM_VERSION;
hdr.rtm_type = action;
hdr.rtm_tableid = kr_state.rtableid;
@@ -2096,6 +2099,11 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = prefix;
iov[iovcnt++].iov_len = sizeof(prefix);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
+
if (memcmp(kroute-nexthop, in6addr_any, sizeof(struct in6_addr))) {
bzero(nexthop, sizeof(nexthop));
@@ -2110,6 +2118,10 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = nexthop;
iov[iovcnt++].iov_len = sizeof(nexthop);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
}
bzero(mask, sizeof(mask));
@@ -2123,6 +2135,10 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = mask;
iov[iovcnt++].iov_len = sizeof(mask);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
if (kroute-labelid) {
bzero(label, sizeof(label));
Re: bgpd fails to install ipv6 routes in kernel routing table
On Mon, Feb 09, 2009 at 10:13:42AM +0100, Falk Brockerhoff - smartTERRA GmbH wrote: Am 09.02.2009 um 09:53 schrieb Claudio Jeker: Please try the attached diff. A general question about diffs like this: will these diffs automatically go to -current in the next couple of days/weeks? Or do I have to apply all these patches by hand? If the diff works it will go into -current. So currently I'm waiting for positive test results and hopefully an ok by henning@ -- :wq Claudio
Common problem with X60 and X61 (was: Fwd: laptop heating due to wpi(4)?)
Sorry for not doing my homework properly... I came across threads on lenovo forums which discuss this issue with the wireless adapter overheating, and it appears to be a common problem across X60 and X61 laptops. The only difference here is -- on Windows XP, the right palmrest heat is bearable, but under OpenBSD, it just gets too hot for comfort. To rephrase, is there something that can be done on OpenBSD that will reduce this heat? (an alternative OS-neutral solution is to use a USB powered fan that cools that right side constantly!) -Amarendra -- Forwarded message -- From: Amarendra Godbole amarendra.godb...@gmail.com Date: Mon, Feb 9, 2009 at 1:05 PM Subject: laptop heating due to wpi(4)? To: OpenBSD general usage list misc@openbsd.org i recently started using intel wireless on my thinkpad x60, through the wpi(4) driver. earlier, i had heating issues, which were resolved by setting hw.setperf to 0, but now i again see my laptop heating up -- especially below my right palm. temperature sensor outputs from sysctl shows: hw.sensors.acpitz0.temp0=60.05 degC (zone temperature) hw.sensors.acpitz1.temp0=60.05 degC (zone temperature) hw.sensors.cpu0.temp0=58.00 degC hw.sensors.wpi0.raw0=155 (temperature 0 - 285) hw.sensors.aps0.temp0=57.00 degC hw.sensors.aps0.temp1=57.00 degC wpi shows 155, which is roughly 68 deg C. is the heating because of wpi? that's what has changed. any pointers to cooling down the laptop will be appreciated. dmesg, if needed, is here http://www.obscure.org/~amunix/tmp/dmesg thanks. -amarendra
Re: bgpd fails to install ipv6 routes in kernel routing table
On Mon, Feb 09, 2009 at 02:22:08AM -0800, patrick keshishian wrote:
On Mon, Feb 9, 2009 at 12:53 AM, Claudio Jeker cje...@diehard.n-r-g.com
wrote:
On a hunch, I tried a 64bit and a 32 bit machine with 1 prefix each.
The 32bit machine adds routes to the kernel without complaint. The
64bit machine complained with send_rtmsg
Arrg. IPv6 is once again broken by design. For some ridiculous reason
struct sockaddr_in6's size is 28 bytes. So IPv6 fucks up alignment on 64 bit
archs. All hail link local addressing and all the crappy workarounds
needed for it.
Maybe it is too late for me to be thinking about this ... but could
you explain the diff below? Unless I'm missing something obvious, it
looks like it changes behavior for non-64bit archs as well.
Hmm. I think your right. I think a different approach would be better.
Will cook up something later today.
--patrick
Please try the attached diff.
--
:wq Claudio
Index: kroute.c
===
RCS file: /cvs/src/usr.sbin/bgpd/kroute.c,v
retrieving revision 1.164
diff -u -p -r1.164 kroute.c
--- kroute.c9 Feb 2009 08:20:11 - 1.164
+++ kroute.c9 Feb 2009 08:49:00 -
@@ -2057,12 +2057,14 @@ retry:
int
send_rt6msg(int fd, int action, struct kroute6 *kroute)
{
- struct ioveciov[5];
+ struct ioveciov[8];
struct rt_msghdrhdr;
struct sockaddr_in6 prefix;
struct sockaddr_in6 nexthop;
struct sockaddr_in6 mask;
struct sockaddr_rtlabel label;
+ chargrmbl[sizeof(long) - (sizeof(prefix)
+ (sizeof(long) - 1))];
int iovcnt = 0;
if (kr_state.fib_sync == 0)
@@ -2070,6 +2072,7 @@ send_rt6msg(int fd, int action, struct k
/* initialize header */
bzero(hdr, sizeof(hdr));
+ bzero(grmbl, sizeof(grmbl));
hdr.rtm_version = RTM_VERSION;
hdr.rtm_type = action;
hdr.rtm_tableid = kr_state.rtableid;
@@ -2096,6 +2099,11 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = prefix;
iov[iovcnt++].iov_len = sizeof(prefix);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
+
if (memcmp(kroute-nexthop, in6addr_any, sizeof(struct in6_addr)))
{
bzero(nexthop, sizeof(nexthop));
@@ -2110,6 +2118,10 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = nexthop;
iov[iovcnt++].iov_len = sizeof(nexthop);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
}
bzero(mask, sizeof(mask));
@@ -2123,6 +2135,10 @@ send_rt6msg(int fd, int action, struct k
/* adjust iovec */
iov[iovcnt].iov_base = mask;
iov[iovcnt++].iov_len = sizeof(mask);
+ /* don't we all love IPv6 */
+ hdr.rtm_msglen += sizeof(grmbl);
+ iov[iovcnt].iov_base = grmbl;
+ iov[iovcnt++].iov_len = sizeof(grmbl);
if (kroute-labelid) {
bzero(label, sizeof(label));
--
:wq Claudio
Re: bgpd fails to install ipv6 routes in kernel routing table
Am 09.02.2009 um 11:23 schrieb Claudio Jeker: If the diff works it will go into -current. So currently I'm waiting for positive test results and hopefully an ok by henning@ Perfect. Thank you (and Henning and all the others), once again, for your incredible and fast support! :wq Claudio Regards, Falk
Re: Common problem with X60 and X61 (was: Fwd: laptop heating due to wpi(4)?)
Maybe you could turn down the power of the device, or put it into powersave mode via ifconfig? Sorry for not doing my homework properly... I came across threads on lenovo forums which discuss this issue with the wireless adapter overheating, and it appears to be a common problem across X60 and X61 laptops. The only difference here is -- on Windows XP, the right palmrest heat is bearable, but under OpenBSD, it just gets too hot for comfort. To rephrase, is there something that can be done on OpenBSD that will reduce this heat? (an alternative OS-neutral solution is to use a USB powered fan that cools that right side constantly!) -Amarendra -- Forwarded message -- From: Amarendra Godbole amarendra.godb...@gmail.com Date: Mon, Feb 9, 2009 at 1:05 PM Subject: laptop heating due to wpi(4)? To: OpenBSD general usage list misc@openbsd.org i recently started using intel wireless on my thinkpad x60, through the wpi(4) driver. earlier, i had heating issues, which were resolved by setting hw.setperf to 0, but now i again see my laptop heating up -- especially below my right palm. temperature sensor outputs from sysctl shows: hw.sensors.acpitz0.temp0=60.05 degC (zone temperature) hw.sensors.acpitz1.temp0=60.05 degC (zone temperature) hw.sensors.cpu0.temp0=58.00 degC hw.sensors.wpi0.raw0=155 (temperature 0 - 285) hw.sensors.aps0.temp0=57.00 degC hw.sensors.aps0.temp1=57.00 degC wpi shows 155, which is roughly 68 deg C. is the heating because of wpi? that's what has changed. any pointers to cooling down the laptop will be appreciated. dmesg, if needed, is here http://www.obscure.org/~amunix/tmp/dmesg thanks. -amarendra
Re: laptop heating due to wpi(4)?
Amarendra Godbole wrote: i recently started using intel wireless on my thinkpad x60, through the wpi(4) driver. earlier, i had heating issues, which were resolved by setting hw.setperf to 0, but now i again see my laptop heating up -- especially below my right palm. temperature sensor outputs from sysctl shows: hw.sensors.acpitz0.temp0=60.05 degC (zone temperature) hw.sensors.acpitz1.temp0=60.05 degC (zone temperature) hw.sensors.cpu0.temp0=58.00 degC hw.sensors.wpi0.raw0=155 (temperature 0 - 285) hw.sensors.aps0.temp0=57.00 degC hw.sensors.aps0.temp1=57.00 degC wpi shows 155, which is roughly 68 deg C. is the heating because of wpi? that's what has changed. any pointers to cooling down the laptop will be appreciated. dmesg, if needed, is here http://www.obscure.org/~amunix/tmp/dmesg - OpenBSD 4.4-current (kernel) #11: Wed Jan 21 07:41:19 IST 2009 r...@zimbu.:/home/amar/site-specific/builds/kernel uh.. what happens if you use a GENERIC snapshot rather than your Franken-kernel? Seeing stuff like that causes people to lose interest really quickly. Setting hw.setperf to 0 isn't a resolution but a burying the problem where you don't see it, for now...or then. I don't think wpi is causing your problem, your system doesn't seem to be managing power properly, you buried the problem by reducing power consumption (and performance). Nick.
Problems with OpenBSD 4.4 amd64 install
I have installed 4.4 on amd64. When I boot up I get a pretty dmesg, followed by: Feb 9 18:52:43 init: cannot stat /etc/login.conf: No such file or directory sh: /etc/rc: No such File or directory Feb 9 18:52:45 init: /etc/pwd.db: No such file or directory Enter pathname of shell or RETURN for sh: After this, I get a shell, but nothing but / seems mounted. I could not save the dmesg. -- Bills adding up? Click here for free information on payday loans. http://tagline.hushmail.com/fc/PnY6qxtVaLAhIQqnqZKXCwIB5X2Ej3z8WP82Jy6H6Mb0SxIj3W8ob/
altq merge
Hello Misc, When the final port altq in pf? And then in 2002 and stretches porting CDNR, the kernel ported in pf no. Here is an excerpt from the log of 16.12.2002 about altq Log message: switchover to pf-based altq. - remove files which are no longer used, or we don't have plans to support in pf in the near future. - remove altq ioctl related stuff. - convert the PRIQ, HFSC and RIO modules to pf-based altq. (these are not enabled in GENERIC, CDNR is not converted yet.) When you fully CDNR transfer? -- Best regards, irix mailto:i...@ukr.net
Re: Problems with OpenBSD 4.4 amd64 install
On Mon, 09 Feb 2009 13:38:47 +0100, auto709563 wrote I have installed 4.4 on amd64. When I boot up I get a pretty dmesg, followed by: Feb 9 18:52:43 init: cannot stat /etc/login.conf: No such file or directory sh: /etc/rc: No such File or directory Feb 9 18:52:45 init: /etc/pwd.db: No such file or directory Enter pathname of shell or RETURN for sh: After this, I get a shell, but nothing but / seems mounted. I could not save the dmesg. Very strange. You have a kernel, and you have init(8), but it looks like looks like /etc is missing or damaged. Did you, by chance, create a separate partition for /etc? That could explain this, as /etc is needed in order to boot multi-user.
Re: Problems with OpenBSD 4.4 amd64 install
I wrote: Very strange. You have a kernel, and you have init(8), but it looks like looks like /etc is missing or damaged. Did you, by chance, create a separate partition for /etc? That could explain this, as /etc is needed in order to boot multi-user. Another reason would be forgetting to install the etc44.tgz fileset, which, along with base44.tgz, is *mandatory*.
Re: bgpd fails to install ipv6 routes in kernel routing table
On Mon, Feb 09, 2009 at 11:43:10AM +0100, Claudio Jeker wrote:
On Mon, Feb 09, 2009 at 02:22:08AM -0800, patrick keshishian wrote:
On Mon, Feb 9, 2009 at 12:53 AM, Claudio Jeker cje...@diehard.n-r-g.com
wrote:
On a hunch, I tried a 64bit and a 32 bit machine with 1 prefix each.
The 32bit machine adds routes to the kernel without complaint. The
64bit machine complained with send_rtmsg
Arrg. IPv6 is once again broken by design. For some ridiculous reason
struct sockaddr_in6's size is 28 bytes. So IPv6 fucks up alignment on 64
bit
archs. All hail link local addressing and all the crappy workarounds
needed for it.
Maybe it is too late for me to be thinking about this ... but could
you explain the diff below? Unless I'm missing something obvious, it
looks like it changes behavior for non-64bit archs as well.
Hmm. I think your right. I think a different approach would be better.
Will cook up something later today.
I think this is better. Just compile tested and no real time to test until
later today.
--
:wq Claudio
Index: kroute.c
===
RCS file: /cvs/src/usr.sbin/bgpd/kroute.c,v
retrieving revision 1.164
diff -u -p -r1.164 kroute.c
--- kroute.c9 Feb 2009 08:20:11 - 1.164
+++ kroute.c9 Feb 2009 12:36:18 -
@@ -1813,8 +1813,8 @@ inet6applymask(struct in6_addr *dest, co
dest-s6_addr[i] = src-s6_addr[i] mask.s6_addr[i];
}
-#defineROUNDUP(a, size)\
-(((a) ((size) - 1)) ? (1 + ((a) | ((size) - 1))) : (a))
+#defineROUNDUP(a) \
+(((a) ((sizeof(long)) - 1)) ? (1 + ((a) | ((sizeof(long)) - 1))) : (a))
void
get_rtaddrs(int addrs, struct sockaddr *sa, struct sockaddr **rti_info)
@@ -1825,7 +1825,7 @@ get_rtaddrs(int addrs, struct sockaddr *
if (addrs (1 i)) {
rti_info[i] = sa;
sa = (struct sockaddr *)((char *)(sa) +
- ROUNDUP(sa-sa_len, sizeof(long)));
+ ROUNDUP(sa-sa_len));
} else
rti_info[i] = NULL;
}
@@ -2059,9 +2059,10 @@ send_rt6msg(int fd, int action, struct k
{
struct ioveciov[5];
struct rt_msghdrhdr;
- struct sockaddr_in6 prefix;
- struct sockaddr_in6 nexthop;
- struct sockaddr_in6 mask;
+ struct pad {
+ struct sockaddr_in6 addr;
+ charpad[sizeof(long)];
+ } prefix, nexthop, mask;
struct sockaddr_rtlabel label;
int iovcnt = 0;
@@ -2086,43 +2087,44 @@ send_rt6msg(int fd, int action, struct k
iov[iovcnt++].iov_len = sizeof(hdr);
bzero(prefix, sizeof(prefix));
- prefix.sin6_len = sizeof(prefix);
- prefix.sin6_family = AF_INET6;
- memcpy(prefix.sin6_addr, kroute-prefix, sizeof(struct in6_addr));
+ prefix.addr.sin6_len = sizeof(struct sockaddr_in6);
+ prefix.addr.sin6_family = AF_INET6;
+ memcpy(prefix.addr.sin6_addr, kroute-prefix,
+ sizeof(struct in6_addr));
/* XXX scope does not matter or? */
/* adjust header */
hdr.rtm_addrs |= RTA_DST;
- hdr.rtm_msglen += sizeof(prefix);
+ hdr.rtm_msglen += ROUNDUP(sizeof(struct sockaddr_in6));
/* adjust iovec */
iov[iovcnt].iov_base = prefix;
- iov[iovcnt++].iov_len = sizeof(prefix);
+ iov[iovcnt++].iov_len = ROUNDUP(sizeof(struct sockaddr_in6));
if (memcmp(kroute-nexthop, in6addr_any, sizeof(struct in6_addr))) {
bzero(nexthop, sizeof(nexthop));
- nexthop.sin6_len = sizeof(nexthop);
- nexthop.sin6_family = AF_INET6;
- memcpy(nexthop.sin6_addr, kroute-nexthop,
+ nexthop.addr.sin6_len = sizeof(struct sockaddr_in6);
+ nexthop.addr.sin6_family = AF_INET6;
+ memcpy(nexthop.addr.sin6_addr, kroute-nexthop,
sizeof(struct in6_addr));
/* adjust header */
hdr.rtm_flags |= RTF_GATEWAY;
hdr.rtm_addrs |= RTA_GATEWAY;
- hdr.rtm_msglen += sizeof(nexthop);
+ hdr.rtm_msglen += ROUNDUP(sizeof(struct sockaddr_in6));
/* adjust iovec */
iov[iovcnt].iov_base = nexthop;
- iov[iovcnt++].iov_len = sizeof(nexthop);
+ iov[iovcnt++].iov_len = ROUNDUP(sizeof(struct sockaddr_in6));
}
bzero(mask, sizeof(mask));
- mask.sin6_len = sizeof(mask);
- mask.sin6_family = AF_INET6;
- memcpy(mask.sin6_addr, prefixlen2mask6(kroute-prefixlen),
+ mask.addr.sin6_len = sizeof(struct sockaddr_in6);
+ mask.addr.sin6_family = AF_INET6;
+ memcpy(mask.addr.sin6_addr, prefixlen2mask6(kroute-prefixlen),
sizeof(struct in6_addr));
/* adjust
wpa2 and osx
hi list, i have a problem with wpa2 and osx. i could connect to the ap if i force it to use wpa1 only. all other wpaprotos gives a : WPA2(PSK,unknown/TKIP,AES/TKIP) while scanning with airport and the association failed. the test cases and dmesg could be found here: http://sumi.thepixelz.com/obsd/wpa-openbsd.txt same tests apply to ral(4) in another soekris4801 with the same results. all this was done with a stock snapshot from ftp.openbsd.org fetched 3h ago. xpsp3 behaves a little bit different, wpaprotos wpa1,wpa2 works wpaprotos wpa2 dose not. could anyone verify this behaviour or do i something completely wrong? regards tim p.s. don't care about wpapsk, this is a test setup only ;) -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
Re: laptop heating due to wpi(4)?
On Mon, Feb 9, 2009 at 5:47 PM, Nick Holland n...@holland-consulting.net wrote: Amarendra Godbole wrote: i recently started using intel wireless on my thinkpad x60, through the wpi(4) driver. earlier, i had heating issues, which were resolved by setting hw.setperf to 0, but now i again see my laptop heating up -- especially below my right palm. temperature sensor outputs from sysctl shows: hw.sensors.acpitz0.temp0=60.05 degC (zone temperature) hw.sensors.acpitz1.temp0=60.05 degC (zone temperature) hw.sensors.cpu0.temp0=58.00 degC hw.sensors.wpi0.raw0=155 (temperature 0 - 285) hw.sensors.aps0.temp0=57.00 degC hw.sensors.aps0.temp1=57.00 degC wpi shows 155, which is roughly 68 deg C. is the heating because of wpi? that's what has changed. any pointers to cooling down the laptop will be appreciated. dmesg, if needed, is here http://www.obscure.org/~amunix/tmp/dmesg - OpenBSD 4.4-current (kernel) #11: Wed Jan 21 07:41:19 IST 2009 r...@zimbu.:/home/amar/site-specific/builds/kernel uh.. what happens if you use a GENERIC snapshot rather than your Franken-kernel? Seeing stuff like that causes people to lose interest really quickly. Ummm, okay - its not the stock kernel, but I had the same heating issue with stock kernel too. To confirm, will try with the stock and then report my findings. Setting hw.setperf to 0 isn't a resolution but a burying the problem where you don't see it, for now...or then. I don't think wpi is causing your problem, your system doesn't seem to be managing power properly, you buried the problem by reducing power consumption (and performance). Agreed. After extensive searching, I came across a lenovo forum thread which indicates that X60 series has the wireless card underneath the right palm-rest, which makes them hot when wireless is being used. Apparently this is a design issue with the X60, and the thread is here: http://forums.lenovo.com/lnv/board/message?board.id=X_Series_Thinkpadsmessage.id=22query.id=204119 Though, as you say, the real problem for OBSD is not addressed. Earlier, the laptop used to heat up considerably even when using wired connection (and wireless being disabled by h/w switch). Once hw.setperf was set to 0, the heating became bearable (I was suggested this workaround on misc@ itself). Since it did the trick, I did not bother, until now when the heating re-surfaced. On Windows XP, the laptop does NOT heat up so much, which means there is something else with OBSD, than merely being an X60 problem. I will investigate further with the stock kernel, and then will post my findings. Thanks. -Amarendra
PF Executive Summary
I need to create an executive summary of pf rules and I remember once seeing a link for a tool that read the rules and gave you back a 'checkpoint(ish)' output .. Anyone know the tool I am talking about or can recommend one? Otherwise I am stuck making stick drawings :)
Re: OpenBSD 4.4-release; Lockup after enabling 2nd NIC; both are Linksys EG1032
Yes. Sorry for the self-contradiction. It's been a long day. Just to be sure, I re-installed Ubuntu, and I'm currently doing a system software upgrade with one NIC, and am logged in over the other NIC and running top. (Plus it sees the onboard NIC, but I don't have anything plugged into that.) John On Sun, Feb 8, 2009 at 5:07 PM, patrick keshishian pkesh...@gmail.com wrote: On Sun, Feb 8, 2009 at 4:51 PM, John Mark Schofield r...@sudosu.net wrote: This is looking to me like a bad slot on the motherboard. Which stinks, as this machine is out of warranty. Anyone have any further troubleshooting suggestions? Didn't you state earlier that you had tried the system with Ubuntu and the two NICs worked flawlessly? On Sat, Feb 7, 2009 at 8:02 PM, John Schofield jschofi...@gmail.com wrote: To further attempt to rule out bad hardware, I installed Linux (Ubuntu 8.10). Both NICs operated flawlessly. (I realize that this is not conclusive, as different OS's can exercise hardware in different ways.) -- --- Got root? http://blog.sudosu.net
Using Manual Rebuild in bioctl(8) for Softraid
I noticed that bioctl(8) now includes the -R flag to kick off a manual rebuild. I upgraded to the current snapshot and setup a test mirror to try out the new feature but I must not understand how the syntax works. Assuming bioctl -R works with softraid(4), can someone tell me how to use it correctly? Here's the initial setup of the array and the steps I took: # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d I built the array following the instructions in softraid(4). Everything works as expected. I copied a few 100 MBs from /usr/src to the new array. I then stopped the array with: # bioctl -d sd4 And then attempted to re-create the array without the last element: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which failed with message: bioctl: BIOCCREATERAID: Invalid argument so I forced it with: # bioctl -C force -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which resulted in this message: softraid0: not all chunks were provided softraid0: can't attach metadata type 0 softraid0: not all chunks were provided softraid0: can't attach metadata type 0 scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total I then mounted the array and copied more data to it after which I again deleted it with: # bioctl -d sd4 And then re-created the array with: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a,/dev/wd5d softraid0 which gave this message: scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total bioctl -i sd4 showed: Volume Status Size Device softraid0 0 Online 500105176064 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a I had to delete and recreate with '-C force' to see all three drives. # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d The array mounts but surely must be dirty or inconsistent since I copied data to the degraded array. So I tried the rebuild command in various ways and never could find parameters that would work: # bioctl -R /dev/wd5d sd4 # bioctl -R 0:2.0 sd4 both gave the error: bioctl: BIOCSETSTATE: Invalid argument How do I initiate a manual rebuild? A couple of other questions: 1) How do I manually fail one element in a softraid mirror? I deleted the array and forcibly recreated it minus one element. a) Is that correct? Do I need to run any consistency check after doing so? 2) Is there a way to add new drive to a running softraid so I can then rebuild using it? 3) In the event of an unclean shutdown, is there any process like `radictl -P all` that should be run? Cheers, --Aaron The kernel is custom in that RAIDFrame is enabled. I'm trying to move a mirrored RAIDFrame device to softraid. Other than that no changes were made. OpenBSD 4.5-beta (GENERIC) #1: Mon Feb 9 00:05:23 CST 2009 r...@home.poffenberger.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (AuthenticAMD 686-class, 512KB L2 cache) 2.01 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,MMX,FXSR,SSE,SSE2,HTT,SSE3,CX16 real mem = 1072193536 (1022MB) avail mem = 1028075520 (980MB) mainbus0 at root bios0 at mainbus0: AT/286+ BIOS, date 11/29/06, BIOS32 rev. 0 @ 0xf1fc0, SMBIOS rev. 2.4 @ 0xf (67 entries) bios0: vendor Phoenix Technologies, LTD version ASUS M2NPV-VM ACPI BIOS Revision 0603 date 11/29/2006 bios0: ASUSTek Computer INC. M2NPV-VM acpi0 at bios0: rev 2 acpi0: tables DSDT FACP HPET MCFG APIC acpi0: wakeup devices HUB0(S5) XVRA(S5) XVRB(S5) XVRC(S5) USB0(S4) USB2(S4) AZAD(S5) MMAC(S5) MMCI(S5) UAR1(S5) UAR2(S5) PS2M(S4) PS2K(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpihpet0 at acpi0: 2500 Hz acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0 (boot processor) cpu0: apic clock running at 200MHz cpu at mainbus0: not configured ioapic0 at mainbus0: apid 2 pa 0xfec0, version 11, 24 pins ioapic0: misconfigured as apic 0, remapped to apid 2 acpiprt0 at acpi0: bus 0 (PCI0) acpiprt1 at acpi0: bus 4 (HUB0) acpicpu0 at acpi0 acpitz0 at acpi0: critical temperature 75 degC acpibtn0 at acpi0: PWRB bios0: ROM list: 0xc/0x1 0xd/0x4000! 0xd4000/0x4800 pci0 at mainbus0 bus 0: configuration mode 1 (bios)
Re: usr.sbin/wake removal
Alexander Yurchenko wrote: On Mon, Feb 09, 2009 at 09:05:13PM +1300, Richard Toohey wrote: On 9/02/2009, at 6:31 PM, Thomas Pfaff wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. My guess would be that I want this 10K util, you want that 7.2K util, Fred wants that 20K util, and every Tom, Dick, and Harry wants their n K ... who gets to make the rules, who gets to administer it, etc.? (Who gets to listen to everyone arguing why this or that should go in?) And guess there may be ramifications for install media? you need wake(8) on install media to wake up your local ftp mirror? I can't say the rational for it. Judging by the reaction to the addition and removal, I guess many would use it too. We can only plea to make the removal reverse, that's all we can do. I can only say many times we are told you can leave xbase for example as hard disk is cheap, well that should apply to this very nice tool too. Looks to me that by the number of quick changes done to it between the changes and the removal by a few different developers and the original OK list of a few more and nice comments about it, looks like many would welcome it too. We can only plea to the power to be to change their mind, but that's about it. What I find harsh is the comments made to Marc on undeadly about his removal. He didn't deserved them by a long shut! I have to say it's hard to understand when it's always about the right tool for the job and this is right. However, be as it may. We can only suggest to have it back, however, it's not up to us, nor do we have a say in it in the end either. We don't know the reason, nor do we need to know. The outcome is sad however for sure and obviously many looks like would use it and welcomed it. But I must admit, it's somewhat difficult to understand the statement about bin being full. May be a security possible issue, but if it does only provide wake, then what's the harm in turning up an already up server. (; Hopefully this could be reconsider? If not, thanks anyway to may be review the removal. Best, Daniel
Re: Using Manual Rebuild in bioctl(8) for Softraid
Unfortunately manual rebuild does not work yet on softraid. I'll add that to the man page. On Mon, Feb 09, 2009 at 10:14:24AM -0600, Aaron Poffenberger wrote: I noticed that bioctl(8) now includes the -R flag to kick off a manual rebuild. I upgraded to the current snapshot and setup a test mirror to try out the new feature but I must not understand how the syntax works. Assuming bioctl -R works with softraid(4), can someone tell me how to use it correctly? Here's the initial setup of the array and the steps I took: # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d I built the array following the instructions in softraid(4). Everything works as expected. I copied a few 100 MBs from /usr/src to the new array. I then stopped the array with: # bioctl -d sd4 And then attempted to re-create the array without the last element: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which failed with message: bioctl: BIOCCREATERAID: Invalid argument so I forced it with: # bioctl -C force -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which resulted in this message: softraid0: not all chunks were provided softraid0: can't attach metadata type 0 softraid0: not all chunks were provided softraid0: can't attach metadata type 0 scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total I then mounted the array and copied more data to it after which I again deleted it with: # bioctl -d sd4 And then re-created the array with: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a,/dev/wd5d softraid0 which gave this message: scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total bioctl -i sd4 showed: Volume Status Size Device softraid0 0 Online 500105176064 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a I had to delete and recreate with '-C force' to see all three drives. # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d The array mounts but surely must be dirty or inconsistent since I copied data to the degraded array. So I tried the rebuild command in various ways and never could find parameters that would work: # bioctl -R /dev/wd5d sd4 # bioctl -R 0:2.0 sd4 both gave the error: bioctl: BIOCSETSTATE: Invalid argument How do I initiate a manual rebuild? A couple of other questions: 1) How do I manually fail one element in a softraid mirror? I deleted the array and forcibly recreated it minus one element. a) Is that correct? Do I need to run any consistency check after doing so? 2) Is there a way to add new drive to a running softraid so I can then rebuild using it? 3) In the event of an unclean shutdown, is there any process like `radictl -P all` that should be run? Cheers, --Aaron The kernel is custom in that RAIDFrame is enabled. I'm trying to move a mirrored RAIDFrame device to softraid. Other than that no changes were made. OpenBSD 4.5-beta (GENERIC) #1: Mon Feb 9 00:05:23 CST 2009 r...@home.poffenberger.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (AuthenticAMD 686-class, 512KB L2 cache) 2.01 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,MMX,FXSR,SSE,SSE2,HTT,SSE3,CX16 real mem = 1072193536 (1022MB) avail mem = 1028075520 (980MB) mainbus0 at root bios0 at mainbus0: AT/286+ BIOS, date 11/29/06, BIOS32 rev. 0 @ 0xf1fc0, SMBIOS rev. 2.4 @ 0xf (67 entries) bios0: vendor Phoenix Technologies, LTD version ASUS M2NPV-VM ACPI BIOS Revision 0603 date 11/29/2006 bios0: ASUSTek Computer INC. M2NPV-VM acpi0 at bios0: rev 2 acpi0: tables DSDT FACP HPET MCFG APIC acpi0: wakeup devices HUB0(S5) XVRA(S5) XVRB(S5) XVRC(S5) USB0(S4) USB2(S4) AZAD(S5) MMAC(S5) MMCI(S5) UAR1(S5) UAR2(S5) PS2M(S4) PS2K(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpihpet0 at acpi0: 2500 Hz acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0 (boot processor) cpu0: apic clock running at 200MHz cpu at mainbus0: not configured ioapic0 at mainbus0: apid 2 pa 0xfec0, version 11, 24 pins ioapic0: misconfigured as apic 0, remapped to apid 2
Re: laptop heating due to wpi(4)?
On Mon, Feb 9, 2009 at 5:47 PM, Nick Holland n...@holland-consulting.net wrote: Amarendra Godbole wrote: i recently started using intel wireless on my thinkpad x60, through the wpi(4) driver. earlier, i had heating issues, which were resolved by setting hw.setperf to 0, but now i again see my laptop heating up -- especially below my right palm. temperature sensor outputs from sysctl shows: hw.sensors.acpitz0.temp0=60.05 degC (zone temperature) hw.sensors.acpitz1.temp0=60.05 degC (zone temperature) hw.sensors.cpu0.temp0=58.00 degC hw.sensors.wpi0.raw0=155 (temperature 0 - 285) hw.sensors.aps0.temp0=57.00 degC hw.sensors.aps0.temp1=57.00 degC wpi shows 155, which is roughly 68 deg C. is the heating because of wpi? that's what has changed. any pointers to cooling down the laptop will be appreciated. dmesg, if needed, is here http://www.obscure.org/~amunix/tmp/dmesg - OpenBSD 4.4-current (kernel) #11: Wed Jan 21 07:41:19 IST 2009 r...@zimbu.:/home/amar/site-specific/builds/kernel uh.. what happens if you use a GENERIC snapshot rather than your Franken-kernel? Seeing stuff like that causes people to lose interest really quickly. Setting hw.setperf to 0 isn't a resolution but a burying the problem where you don't see it, for now...or then. I don't think wpi is causing your problem, your system doesn't seem to be managing power properly, you buried the problem by reducing power consumption (and performance). [...] With a stock kernel now: a. when hw.setperf is set to 100, and wireless is on hw.sensors.acpitz0.temp0=78.05 degC (zone temperature) hw.sensors.acpitz1.temp0=76.05 degC (zone temperature) hw.sensors.cpu0.temp0=76.00 degC hw.sensors.wpi0.raw0=154 (temperature 0 - 285) hw.sensors.aps0.temp0=56.00 degC hw.sensors.aps0.temp1=56.00 degC b. when hw.setperf is set to 0, and wireless is on hw.sensors.acpitz0.temp0=53.05 degC (zone temperature) hw.sensors.acpitz1.temp0=52.05 degC (zone temperature) hw.sensors.cpu0.temp0=52.00 degC hw.sensors.wpi0.raw0=146 (temperature 0 - 285) hw.sensors.aps0.temp0=54.00 degC hw.sensors.aps0.temp1=54.00 degC setperf from 100 to 0 decreases the CPU temperature from 76 degC to 52, while there is no appreciable change in that of wpi0, from 154 to 146. For both readings, machine was idle with no human-user activity. The relevant dmesg is put up here http://www.obscure.org/~amunix/tmp/dmesg.bsd.mp.4.4 Any pointers to troubleshoot this issue appreciated. Thanks in advance! -Amarendra
Re: usr.sbin/wake removal
On Mon, Feb 09, 2009 at 09:05:13PM +1300, Richard Toohey wrote: On 9/02/2009, at 6:31 PM, Thomas Pfaff wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. My guess would be that I want this 10K util, you want that 7.2K util, Fred wants that 20K util, and every Tom, Dick, and Harry wants their n K ... who gets to make the rules, who gets to administer it, etc.? (Who gets to listen to everyone arguing why this or that should go in?) And guess there may be ramifications for install media? If there is no room in base, it would be nice to have it in ports. Or is there something else in ports already that does the same thing? I've found wake extremely useful for turning on remote desktop computers from the Soekris firewall rather than leaving them on all the time.
Re: usr.sbin/wake removal
I'd gladly trade look(1) for wake(8). That's almost 8k right there. On 2/9/09, Emilio Perea epe...@walkereng.com wrote: On Mon, Feb 09, 2009 at 09:05:13PM +1300, Richard Toohey wrote: On 9/02/2009, at 6:31 PM, Thomas Pfaff wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. My guess would be that I want this 10K util, you want that 7.2K util, Fred wants that 20K util, and every Tom, Dick, and Harry wants their n K ... who gets to make the rules, who gets to administer it, etc.? (Who gets to listen to everyone arguing why this or that should go in?) And guess there may be ramifications for install media? If there is no room in base, it would be nice to have it in ports. Or is there something else in ports already that does the same thing? I've found wake extremely useful for turning on remote desktop computers from the Soekris firewall rather than leaving them on all the time.
Re: Using Manual Rebuild in bioctl(8) for Softraid
On Feb 9, 2009, at 10:38, Marco Peereboom wrote: Unfortunately manual rebuild does not work yet on softraid. I'll add that to the man page. Thanks. So is the current rebuild process for a failed drive in softraid to build a new array and copy the data from the degraded array to the new array? Cheers, Aaron On Mon, Feb 09, 2009 at 10:14:24AM -0600, Aaron Poffenberger wrote: I noticed that bioctl(8) now includes the -R flag to kick off a manual rebuild. I upgraded to the current snapshot and setup a test mirror to try out the new feature but I must not understand how the syntax works. Assuming bioctl -R works with softraid(4), can someone tell me how to use it correctly? Here's the initial setup of the array and the steps I took: # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d I built the array following the instructions in softraid(4). Everything works as expected. I copied a few 100 MBs from /usr/src to the new array. I then stopped the array with: # bioctl -d sd4 And then attempted to re-create the array without the last element: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which failed with message: bioctl: BIOCCREATERAID: Invalid argument so I forced it with: # bioctl -C force -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which resulted in this message: softraid0: not all chunks were provided softraid0: can't attach metadata type 0 softraid0: not all chunks were provided softraid0: can't attach metadata type 0 scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total I then mounted the array and copied more data to it after which I again deleted it with: # bioctl -d sd4 And then re-created the array with: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a,/dev/wd5d softraid0 which gave this message: scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total bioctl -i sd4 showed: Volume Status Size Device softraid0 0 Online 500105176064 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a I had to delete and recreate with '-C force' to see all three drives. # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d The array mounts but surely must be dirty or inconsistent since I copied data to the degraded array. So I tried the rebuild command in various ways and never could find parameters that would work: # bioctl -R /dev/wd5d sd4 # bioctl -R 0:2.0 sd4 both gave the error: bioctl: BIOCSETSTATE: Invalid argument How do I initiate a manual rebuild? A couple of other questions: 1) How do I manually fail one element in a softraid mirror? I deleted the array and forcibly recreated it minus one element. a) Is that correct? Do I need to run any consistency check after doing so? 2) Is there a way to add new drive to a running softraid so I can then rebuild using it? 3) In the event of an unclean shutdown, is there any process like `radictl -P all` that should be run? Cheers, --Aaron The kernel is custom in that RAIDFrame is enabled. I'm trying to move a mirrored RAIDFrame device to softraid. Other than that no changes were made. OpenBSD 4.5-beta (GENERIC) #1: Mon Feb 9 00:05:23 CST 2009 r...@home.poffenberger.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (AuthenticAMD 686-class, 512KB L2 cache) 2.01 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,MMX,FXSR,SSE,SSE2,HTT,SSE3,CX16 real mem = 1072193536 (1022MB) avail mem = 1028075520 (980MB) mainbus0 at root bios0 at mainbus0: AT/286+ BIOS, date 11/29/06, BIOS32 rev. 0 @ 0xf1fc0, SMBIOS rev. 2.4 @ 0xf (67 entries) bios0: vendor Phoenix Technologies, LTD version ASUS M2NPV-VM ACPI BIOS Revision 0603 date 11/29/2006 bios0: ASUSTek Computer INC. M2NPV-VM acpi0 at bios0: rev 2 acpi0: tables DSDT FACP HPET MCFG APIC acpi0: wakeup devices HUB0(S5) XVRA(S5) XVRB(S5) XVRC(S5) USB0(S4) USB2(S4) AZAD(S5) MMAC(S5) MMCI(S5) UAR1(S5) UAR2(S5) PS2M(S4) PS2K(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpihpet0 at acpi0: 2500 Hz acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0 (boot processor) cpu0: apic clock running at 200MHz cpu
С наилучшими пожеланиями!
http://uralweb.ru - dna`b| b hgap`mmne!
From: maksimfi...@uralweb.ru
Subject:
To: root_at_localh...@jabber.ru,
X-Mailer: CommuniGate Pro WebUser Interface v.4.3.12
Date: Mon, 09 Feb 2009 22:09:33 +0500
Message-ID: web-62152...@uralweb.ru
MIME-Version: 1.0
X-Security: message sanitized on shear.ucar.edu See
http://www.impsec.org/email-tools/sanitizer-intro.html for details. $Revision:
1.147 $Date: 2004-10-02 11:16:26-07
Content-Type: text/plain; charset=us-ascii
X-Converted-To-Plain-Text: from multipart/mixed by demime 1.01d
X-Converted-To-Plain-Text: Alternative section used was text/html
Gdp`bqrbsire, lem gnbsr L`jqhl. Onf`ksiqr`, on bnglnfmnqrh, dnwhr`ire
}rn ohq|ln dn jnmv`. Opedk`c`~, B`l bnglnfmnqr| dnonkmhrek|mncn
g`p`anrj`! B{ qopnqhre, onwels opedk`c`~ h onwels hlemmn B`l? Nrbews
weqrmn lme, aeg p`gmhv{, jnls opedk`c`r|, opnqrn m`xek qq{kjs m` B`x
}kejrpnmm{i `dpeq m` q`ire n tphk`mqe (Tphk`mq (`mck. free lance
qbnandmne jno|e) sd`k8mm` p`anr`). H ondsl`k, wrn hlemmn B`q, }rn
lnckn a{ g`hmrepeqnb`r|. Oph fek`mhh h `jrhbmnl sw`qrhh b opncp`lle
(ophlepmn 1 hkh 2 w`q` b dem|) B{ qlnfere g`p`anr`r| dn 500 h ankee r{qw
psakei b onqkeds~yhe 3 leqv`. H }rn me xsrj`! ]rn deiqrbhrek|mn
p`anr`er! B }rnl mer mhwecn qknfmncn! _ b }rnl sfe saedhkq hm`we, me
rp`rhk a{ qbne dp`cnvemmne bpel m` bqe }rh p`qq{kjh! P`qqj`fs melmncn n
qeae. Lme 35 ker, p`anr`~ lemedfepnl nrdek` opnd`f b jnlo`mhh
opnhgbndhreke qrpnil`reph`knb. J`j me keqrmn gbswhr, qwhr`~ qea, d`kejn
me ckso{l weknbejnl. S lem b{qxee csl`mhr`pmne nap`gnb`mhe. K~ak~
khrep`rsps. J`j Pnqqhiqjs~, r`j h g`psaefms~ jk`qqhjs h me rnk|jn.
M`ophlep, hg hmnqrp`mm{u ohq`rekei, onkswhk med`bmn sdnbnk|qrbhe, nr
opnwremh jmhc +Bnkub; T`skg`, +Jnk` Ap~m|nm; Pnlem` Pnkk`m`,
boew`rkhrek|mni h nazelmni +Q`ch n Tnpq`ir`u; Cnkqsnpqh h +...@lephjh;
J`tjh. Qeiw`q o{r`~q| nqhkhr| Slaeprn ]jn. Hg pnqqhiqjhu a{k g`hmrphcnb`m
Dnbk`rnb{l, @jqemnb{l, ms h jsd` fe aeg Dnqrnebqjncn, Rnkqrncn, Askc`jnb`
h lmnchu dpschu opejp`qm{u jk`qqhjnb. B ankee lnkndnl bngp`qre
hmrepeqnb`kq thknqnthei. G`mhl`~q| thrmeqnl, hcp`~ b ahk|pd h remmhq, `
r`jfe k~ak~ p{a`kjs. _ cksanjn op`bnqk`bm{i weknbej h qwhr`~ wrn ck`bmne
b fhgm| meqrh k~dl dnapne, bewmne h k~ahr| akhfmhu qbnhu. Rn eqr|,
na{wm{i weknbej. Lni njk`d m` p`anre qnqr`bker nr 35 dn 50 r{q. psa. b
leqv. Nm g`bhqhr nr opnvemrnb q opnd`f. Onqke m`w`k` p`anr{ q
opncp`llni, jnrnps~, b`l opedk`c`~, opnxkn 2 leqv` h g`p`anr`k onwrh
150 r{q.psa.!!! _ q`l b xnje!!! Bqe }rn, pexhk m`ohq`r|, qosqr 2
leqv` sw`qrh b }rni opncp`lle, wrna{ ankee dnundwhbn nazqmhr| nayhi
ql{qk. Bqe ondpnamnqrh b{ m`idere bn bknfemhh! Opnxs me sd`kr| }rn
ohq|ln, qp`gs onk`c` wrn }rn qo`l qeiw`q b }rn bephr|q q rpsdnl mn
opnwrhre bknfemm{i t`ik h }rn ohq|ln eye p`g h bqe onilere! H onlmhre
}rn p...@k\m[e dem|ch, jnrnp{e lnfmn onrpnc`r| h onrp`rhr| m` bqe, wrn B`l
g`ak`cnp`qqsdhrq! Sd`wmncn B`l dm h unpnxecn m`qrpnemh! L`jqhl.
http://uralweb.ru - dna`b| b hgap`mmne!
[demime 1.01d removed an attachment of type APPLICATION/DEFANGED]
Re: laptop heating due to wpi(4)?
* Amarendra Godbole amarendra.godb...@gmail.com [090209 09:30]: On Mon, Feb 9, 2009 at 5:47 PM, Nick Holland n...@holland-consulting.net wrote: Amarendra Godbole wrote: i recently started using intel wireless on my thinkpad x60, through the wpi(4) driver. earlier, i had heating issues, which were resolved by setting hw.setperf to 0, but now i again see my laptop heating up -- especially below my right palm. temperature sensor outputs from sysctl shows: hw.sensors.acpitz0.temp0=60.05 degC (zone temperature) hw.sensors.acpitz1.temp0=60.05 degC (zone temperature) hw.sensors.cpu0.temp0=58.00 degC hw.sensors.wpi0.raw0=155 (temperature 0 - 285) hw.sensors.aps0.temp0=57.00 degC hw.sensors.aps0.temp1=57.00 degC wpi shows 155, which is roughly 68 deg C. is the heating because of wpi? that's what has changed. any pointers to cooling down the laptop will be appreciated. dmesg, if needed, is here http://www.obscure.org/~amunix/tmp/dmesg - OpenBSD 4.4-current (kernel) #11: Wed Jan 21 07:41:19 IST 2009 r...@zimbu.:/home/amar/site-specific/builds/kernel uh.. what happens if you use a GENERIC snapshot rather than your Franken-kernel? Seeing stuff like that causes people to lose interest really quickly. Ummm, okay - its not the stock kernel, but I had the same heating issue with stock kernel too. To confirm, will try with the stock and then report my findings. Setting hw.setperf to 0 isn't a resolution but a burying the problem where you don't see it, for now...or then. I don't think wpi is causing your problem, your system doesn't seem to be managing power properly, you buried the problem by reducing power consumption (and performance). Agreed. After extensive searching, I came across a lenovo forum thread which indicates that X60 series has the wireless card underneath the right palm-rest, which makes them hot when wireless is being used. Apparently this is a design issue with the X60, and the thread is here: http://forums.lenovo.com/lnv/board/message?board.id=X_Series_Thinkpadsmessage.id=22query.id=204119 Though, as you say, the real problem for OBSD is not addressed. Earlier, the laptop used to heat up considerably even when using wired connection (and wireless being disabled by h/w switch). Once hw.setperf was set to 0, the heating became bearable (I was suggested this workaround on misc@ itself). Since it did the trick, I did not bother, until now when the heating re-surfaced. On Windows XP, the laptop does NOT heat up so much, which means there is something else with OBSD, than merely being an X60 problem. I will investigate further with the stock kernel, and then will post my findings. Thanks. -Amarendra I can attest to the heat issue tracking i386 -current ever since installing on my x60s year(s) ago. The right palm area does get noticably warmer than when running Windows. It just never became a big enough bother for me to do anything about. Jim
altq merge
irix wrote: Hello Misc, When the final port altq in pf? And then in 2002 and stretches porting CDNR, the \ kernel ported in pf no. Here is an excerpt from the log of 16.12.2002 about altq Log message: switchover to pf-based altq. - remove files which are no longer used, or we don't have plans to support in pf in the near future. - remove altq ioctl related stuff. - convert the PRIQ, HFSC and RIO modules to pf-based altq. (these are not enabled in GENERIC, CDNR is not converted yet.) When you fully CDNR transfer? http://unix.derkeiler.com/Mailing-Lists/FreeBSD/net/2004-10/0137.html man 5 pf.conf is also useful. so is OpenBSD FAQ. Alexey
Problem of recognizing sangoma A102 2006 in openbsd 4.4
Hi, I have Soekris net4801. I have installed openbsd 4.4 on it. net 4801 has sangoma A102 2006(AFT series)card. Everything works fine except i got message during booting Vendor Sangoma, Unknown Product 0x0040 (Class network subclass miscellaneous, rev 0x00) at pci0 dev 10 function 0 not configured Can anybody tell me, what should i do , so that openbsd can recognize sangoma?)
Re: Using Manual Rebuild in bioctl(8) for Softraid
Dump the data to a backup; recreate softraid disk; restore data from dump. On Mon, Feb 09, 2009 at 11:38:23AM -0600, Aaron Poffenberger wrote: On Feb 9, 2009, at 10:38, Marco Peereboom wrote: Unfortunately manual rebuild does not work yet on softraid. I'll add that to the man page. Thanks. So is the current rebuild process for a failed drive in softraid to build a new array and copy the data from the degraded array to the new array? Cheers, Aaron On Mon, Feb 09, 2009 at 10:14:24AM -0600, Aaron Poffenberger wrote: I noticed that bioctl(8) now includes the -R flag to kick off a manual rebuild. I upgraded to the current snapshot and setup a test mirror to try out the new feature but I must not understand how the syntax works. Assuming bioctl -R works with softraid(4), can someone tell me how to use it correctly? Here's the initial setup of the array and the steps I took: # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d I built the array following the instructions in softraid(4). Everything works as expected. I copied a few 100 MBs from /usr/src to the new array. I then stopped the array with: # bioctl -d sd4 And then attempted to re-create the array without the last element: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which failed with message: bioctl: BIOCCREATERAID: Invalid argument so I forced it with: # bioctl -C force -c 1 -l /dev/wd5a,/dev/wd6a softraid0 which resulted in this message: softraid0: not all chunks were provided softraid0: can't attach metadata type 0 softraid0: not all chunks were provided softraid0: can't attach metadata type 0 scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total I then mounted the array and copied more data to it after which I again deleted it with: # bioctl -d sd4 And then re-created the array with: # bioctl -c 1 -l /dev/wd5a,/dev/wd6a,/dev/wd5d softraid0 which gave this message: scsibus2 at softraid0: 1 targets, initiator 1 sd4 at scsibus2 targ 0 lun 0: OPENBSD, SR RAID 1, 003 SCSI2 0/direct fixed sd4: 476937MB, 512 bytes/sec, 976767923 sec total bioctl -i sd4 showed: Volume Status Size Device softraid0 0 Online 500105176064 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a I had to delete and recreate with '-C force' to see all three drives. # bioctl -i sd4 Volume Status Size Device softraid0 0 Online 500099595776 sd4 RAID1 0 Online 500105176064 0:0.0 noencl wd5a 1 Online 500105176064 0:1.0 noencl wd6a 2 Online 500099595776 0:2.0 noencl wd5d The array mounts but surely must be dirty or inconsistent since I copied data to the degraded array. So I tried the rebuild command in various ways and never could find parameters that would work: # bioctl -R /dev/wd5d sd4 # bioctl -R 0:2.0 sd4 both gave the error: bioctl: BIOCSETSTATE: Invalid argument How do I initiate a manual rebuild? A couple of other questions: 1) How do I manually fail one element in a softraid mirror? I deleted the array and forcibly recreated it minus one element. a) Is that correct? Do I need to run any consistency check after doing so? 2) Is there a way to add new drive to a running softraid so I can then rebuild using it? 3) In the event of an unclean shutdown, is there any process like `radictl -P all` that should be run? Cheers, --Aaron The kernel is custom in that RAIDFrame is enabled. I'm trying to move a mirrored RAIDFrame device to softraid. Other than that no changes were made. OpenBSD 4.5-beta (GENERIC) #1: Mon Feb 9 00:05:23 CST 2009 r...@home.poffenberger.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (AuthenticAMD 686-class, 512KB L2 cache) 2.01 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,MMX,FXSR,SSE,SSE2,HTT,SSE3,CX16 real mem = 1072193536 (1022MB) avail mem = 1028075520 (980MB) mainbus0 at root bios0 at mainbus0: AT/286+ BIOS, date 11/29/06, BIOS32 rev. 0 @ 0xf1fc0, SMBIOS rev. 2.4 @ 0xf (67 entries) bios0: vendor Phoenix Technologies, LTD version ASUS M2NPV-VM ACPI BIOS Revision 0603 date 11/29/2006 bios0: ASUSTek Computer INC. M2NPV-VM acpi0 at bios0: rev 2 acpi0: tables DSDT FACP HPET MCFG APIC acpi0: wakeup devices HUB0(S5) XVRA(S5) XVRB(S5) XVRC(S5) USB0(S4) USB2(S4) AZAD(S5) MMAC(S5) MMCI(S5)
Re: usr.sbin/wake removal
On Feb 8, 2009, at 9:31 PM, Thomas Pfaff wrote: On Sun, 8 Feb 2009 15:53:01 -0700 (MST) Marc Balmer mbal...@cvs.openbsd.org wrote: CVSROOT:/cvs Module name:src Changes by: mbal...@cvs.openbsd.org 2009/02/08 15:53:01 Removed files: usr.sbin/wake : Makefile wake.8 wake.c Log message: Remove wake(8). The bin directories are full, no new commands to be added. I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. I'm curious about this as well. What sort of resource limitation is being hit here? -- bk
relayd ssl cluster + virtual domains
I'm trying to set up relayd to use as an https proxy to a cluster of virtual domains. I've read https://calomel.org/relayd.html and gone through the manpages and do not see how to send a different cert depending on the domain requested. I'm ok with the no encryption between relayd and the cluster as the colomel site describes. I've gotten my sample site up and running but now need to add virtualhosts to that. Will relayd be able to handle this? Or should I be using gnutls for this? Thanks, -Bryan
Re: usr.sbin/wake removal
On Mon, Feb 09, 2009 at 11:00:03AM -0800, Brian Keefer wrote: On Feb 8, 2009, at 9:31 PM, Thomas Pfaff wrote: On Sun, 8 Feb 2009 15:53:01 -0700 (MST) Marc Balmer mbal...@cvs.openbsd.org wrote: CVSROOT:/cvs Module name:src Changes by: mbal...@cvs.openbsd.org 2009/02/08 15:53:01 Removed files: usr.sbin/wake : Makefile wake.8 wake.c Log message: Remove wake(8). The bin directories are full, no new commands to be added. I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. I'm curious about this as well. What sort of resource limitation is being hit here? EINTERNALPOLITICS -- bk
Re: usr.sbin/wake removal
On Monday 09 February 2009 11:15:25 Emilio Perea wrote: On Mon, Feb 09, 2009 at 09:05:13PM +1300, Richard Toohey wrote: On 9/02/2009, at 6:31 PM, Thomas Pfaff wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. Thanks. My guess would be that I want this 10K util, you want that 7.2K util, Fred wants that 20K util, and every Tom, Dick, and Harry wants their n K ... who gets to make the rules, who gets to administer it, etc.? (Who gets to listen to everyone arguing why this or that should go in?) And guess there may be ramifications for install media? If there is no room in base, it would be nice to have it in ports. Or is there something else in ports already that does the same thing? I've found wake extremely useful for turning on remote desktop computers from the Soekris firewall rather than leaving them on all the time. /usr/ports/net/wol has existed for some time now. I like the idea of a builtin wake more though. You can always keep a copy of it and build it yourself. Thats what I've done. --STeve Andre'
Re: Thinkpad R61 support
On Monday 09 February 2009 01:59:56 Michiel van Baak wrote: On 01:04, Mon 09 Feb 09, Marc Espie wrote: On Sun, Feb 08, 2009 at 08:37:43AM -0500, Kenneth R Westerback wrote: On Sat, Feb 07, 2009 at 07:39:24PM -0200, Christiano Farina Haesbaert wrote: Hello there, I'm considering buying a thinkpad R61, if someone has any information on the hardware support for it I would appreciate. Best Regards. -- Christiano Farina Haesbaert We bought a bunch of R61's at work and had nothing but trouble with them, especially the wireless. But this is with Windows and not OpenBSD. They also weigh a ton. The wireless works under OpenBSD, but it loses network once in a while. The best fix so far is some ifconfig iwn0 down; dhclient iwn0 That makes it work again... ipw in the T61p has the same. Once a week or something. I think these iwn problems are machine specific. I have a W500 ThinkPad, and iwn0 is rock stable here. As long as I have a signal, I have a connection, and, its more sensitive than other laptops. --STeve Andre'
Re: usr.sbin/wake removal
On Mon, Feb 9, 2009 at 5:15 PM, Emilio Perea epe...@walkereng.com wrote: If there is no room in base, it would be nice to have it in ports. There's no more room in ports either. Landry
Re: relayd ssl cluster + virtual domains
On 2009-02-09, Bryan Irvine sparcta...@gmail.com wrote: I've read https://calomel.org/relayd.html and gone through the manpages and do not see how to send a different cert depending on the domain requested. Assuming you mean name-based virtual hosts using the Host header from the HTTP request: you can't. The Host header is sent *after* SSL is negotiated.
Re: usr.sbin/wake removal
On Mon, Feb 9, 2009 at 12:31 AM, Thomas Pfaff tpf...@tp76.info wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. I'm kinda amazed at the hoopla over this. Last week a wake on lan utility was like the only possible feature not being requested, you didn't even know you wanted it, and now a week later it's like people can't live without it. Yeah, it's handy, but if you survived 10 years without it, I think you can get by a little longer. Things usually happen for a reason (though I've no idea what they are here), but at least wait until the dust settles before starting the inquisition. Nothing speeds progress like a dozen how come?s at every step.
Re: Problem of recognizing sangoma A102 2006 in openbsd 4.4
Prakshep Dineshchandra Patel wrote: Hi, I have Soekris net4801. I have installed openbsd 4.4 on it. net 4801 has sangoma A102 2006(AFT series)card. Everything works fine except i got message during booting Vendor Sangoma, Unknown Product 0x0040 (Class network subclass miscellaneous, rev 0x00) at pci0 dev 10 function 0 not configured Can anybody tell me, what should i do , so that openbsd can recognize sangoma?) Hello Prakshep, You may consider looking at the following files /usr/src/sys/dev/pci/pcidevs.h /usr/src/sys/dev/pci/if_sandrv.c to see if your particular PCI card version is supported by the san(4) driver by adding the product code for your card into the driver. Then again, as Sangoma wrote the driver, you might try contacting them for more information. And searching for sangoma in the archives provides the following threads: http://marc.info/?l=openbsd-miscm=118245939832197w=2 http://marc.info/?l=openbsd-miscm=119403349716792w=2 -Tico
Re: relayd ssl cluster + virtual domains
On 11:08, Mon 09 Feb 09, Bryan Irvine wrote: I'm trying to set up relayd to use as an https proxy to a cluster of virtual domains. I've read https://calomel.org/relayd.html and gone through the manpages and do not see how to send a different cert depending on the domain requested. I'm ok with the no encryption between relayd and the cluster as the colomel site describes. I've gotten my sample site up and running but now need to add virtualhosts to that. Will relayd be able to handle this? Or should I be using gnutls for this? You should use a different ip address for every ssl cert. Name-based virtual hosting is not supported by HTTPS because encrytion is setup before the host headers are sent back-n-forth. This is the case with relayd, apache, thttpd, $your_fav_httpd Yes, apache only logs a warning in the logs and tries some tricks to get around this but that's ugly. -- Michiel van Baak mich...@vanbaak.eu http://michiel.vanbaak.eu GnuPG key: http://pgp.mit.edu:11371/pks/lookup?op=getsearch=0x71C946BD Why is it drug addicts and computer aficionados are both called users?
relayd ssl to ssl not working. Sends http request to https port
I posted a couple weeks ago about my relayd configuration and how it seemed
that it was not relaying traffic. Since then I have been trying to simply
the configuration and make *something* work. I was successful in getting
relayd to listen on port 80 and forward traffic to a group of other web
servers on port 80. However, I haven't been able to do anything more
complicated than that.
Right now I am trying to listen on port 443 for incoming connections and
relay them to a group of web servers that are listening on port 443. Most
of the time, nothing happens. It just seems to hang there. However, I did
manage to get a useful error from a web server the other day.
Your browser sent a request that this server could not understand.
Reason: You're speaking plain HTTP to an SSL-enabled server port.
Instead use the HTTPS scheme to access this URL, please.
Is there something in my configuration file that I need to specify to ensure
that https requests are sent to the servers? I've looked at a few examples
online and I haven't seen anything that fits the bill. Here is my
relayd.conf file
table ssl_server { www.mnsu.edu, secure.mnsu.edu }
web_port=80
ssl_port=443
bge0_ip=134.29.32.88
interval 10
timeout 200
prefork 5
log updates
http protocol httpfilter {
# TCP Performance options
tcp { nodelay, sack, socket buffer 65536, backlog 100 }
# Return HTTP/HTML error pages
return error
# allow logging of remote client ips to internal web servers
header append $REMOTE_ADDR to X-Forwarded-For
# Set keep alive timeout to global timeout
header change Keep-Alive to $TIMEOUT
# Close connection upon receipt
header change Connection to close
# Anonymize webservers name/type
response header change Server to Something
# SSL options
ssl { sslv3, tlsv1, ciphers HIGH:!ADH, no sslv2 }
}
relay web_proxy {
listen on $bge0_ip port $ssl_port ssl
protocol httpfilter
forward to ssl_server port $ssl_port mode loadbalance check https /
code 200
}
Re: altq merge
Hello Misc, In that freebsd list tell as in 2002 in OpenBSD list will merge. The options ALTQ_CDNR is a dummy at the moment. It introduces a function pointer in ip_input() that can be used as conditioner hook, but it is not used at the moment. There are plans to resurrect the conditioner, but it is not yet clear how and where. It might be a function of pf in the future. But since 2002 it has been 6 years and will merge and stand still. When the CDNR will merge in pf ??? -- Best regards, irix mailto:i...@ukr.net
Re: relayd ssl to ssl not working. Sends http request to https port
On 2009-02-09, kevin thompson kevin.david.thomp...@gmail.com wrote:
Is there something in my configuration file that I need to specify to ensure
that https requests are sent to the servers? I've looked at a few examples
online and I haven't seen anything that fits the bill. Here is my
relayd.conf file
basically it looks like you want to decrypt, adjust the headers,
and then re-encrypt to the server.
relayd doesn't have this feature (mitm mode? :-)
it could probably be added as an option to forward to for a
relay, but this would bring some questions about how to handle
invalid certificates at the backend server, etc... (and without
safe ways to handle that, you might as well keep the cleartext
to the backend).
with what's currently available in relayd, you would have to
use a plain TCP relay for HTTPS.
table ssl_server { www.mnsu.edu, secure.mnsu.edu }
web_port=80
ssl_port=443
bge0_ip=134.29.32.88
interval 10
timeout 200
prefork 5
log updates
http protocol httpfilter {
# TCP Performance options
tcp { nodelay, sack, socket buffer 65536, backlog 100 }
# Return HTTP/HTML error pages
return error
# allow logging of remote client ips to internal web servers
header append $REMOTE_ADDR to X-Forwarded-For
# Set keep alive timeout to global timeout
header change Keep-Alive to $TIMEOUT
# Close connection upon receipt
header change Connection to close
# Anonymize webservers name/type
response header change Server to Something
# SSL options
ssl { sslv3, tlsv1, ciphers HIGH:!ADH, no sslv2 }
}
relay web_proxy {
listen on $bge0_ip port $ssl_port ssl
protocol httpfilter
forward to ssl_server port $ssl_port mode loadbalance check https /
code 200
}
Re: Multihead console
On Sat, Feb 7, 2009 at 3:09 AM, Incognito incogn...@electrostatik.org wrote: Is it possible to associate virtual terminals to multiple video cards and monitors? So when you start ctrl-alt'ing through terminals you'll jump to different physical screens. Couldn't find any conclusive info after some googling and list digging. (did read wsdisplay(4), wscons(4), wsconscfg(8) and wsconsctl(8)). No. (At least not without writing code to support that both in the kernel and userland...) -- Matthieu Herrb
Re: Multihead console
No. (At least not without writing code to support that both in the kernel and userland...) are you aware of any *BSD that does support multiple consoles? Sam Fourman Jr. Fourman Networks
Re: altq merge
On 2009-02-09, irix i...@ukr.net wrote: Hello Misc, In that freebsd list tell as in 2002 in OpenBSD list will merge. The options ALTQ_CDNR is a dummy at the moment. It introduces a function pointer in ip_input() that can be used as conditioner hook, but it is not used at the moment. There are plans to resurrect the conditioner, but it is not yet clear how and where. It might be a function of pf in the future. But since 2002 it has been 6 years and will merge and stand still. When the CDNR will merge in pf ??? when someone has enough need for it that they do the work.
Re: relayd ssl to ssl not working. Sends http request to https port
I see what you're saying. I was wondering how MITM would work too, and I
just assumed there was some magic built into relayd.
I don't actually want to modify the headers and stuff, I really just want to
forward the traffic like a load balancer. I just followed the example for
setting up an http relay and assumed that setting up an https relay was
almost the same. I'll try using a regular tcp relay. Thank you.
Kevin
On Mon, Feb 9, 2009 at 3:15 PM, Stuart Henderson s...@spacehopper.orgwrote:
On 2009-02-09, kevin thompson kevin.david.thomp...@gmail.com wrote:
Is there something in my configuration file that I need to specify to
ensure
that https requests are sent to the servers? I've looked at a few
examples
online and I haven't seen anything that fits the bill. Here is my
relayd.conf file
basically it looks like you want to decrypt, adjust the headers,
and then re-encrypt to the server.
relayd doesn't have this feature (mitm mode? :-)
it could probably be added as an option to forward to for a
relay, but this would bring some questions about how to handle
invalid certificates at the backend server, etc... (and without
safe ways to handle that, you might as well keep the cleartext
to the backend).
with what's currently available in relayd, you would have to
use a plain TCP relay for HTTPS.
table ssl_server { www.mnsu.edu, secure.mnsu.edu }
web_port=80
ssl_port=443
bge0_ip=134.29.32.88
interval 10
timeout 200
prefork 5
log updates
http protocol httpfilter {
# TCP Performance options
tcp { nodelay, sack, socket buffer 65536, backlog 100 }
# Return HTTP/HTML error pages
return error
# allow logging of remote client ips to internal web servers
header append $REMOTE_ADDR to X-Forwarded-For
# Set keep alive timeout to global timeout
header change Keep-Alive to $TIMEOUT
# Close connection upon receipt
header change Connection to close
# Anonymize webservers name/type
response header change Server to Something
# SSL options
ssl { sslv3, tlsv1, ciphers HIGH:!ADH, no sslv2 }
}
relay web_proxy {
listen on $bge0_ip port $ssl_port ssl
protocol httpfilter
forward to ssl_server port $ssl_port mode loadbalance check https
/
code 200
}
Re: Segfault under MS Virtual Server 2005 R2 SP1 Enterprise Edition
The Error message was unknown error, help or very nearly. However, as the system that it was to be installed on is a production server, my 'boss' decided that we shouldn't be attempting to do development work of installing SaMBa on a VM to implement Active Directory Single-Sign-On. When playing with your own systems at home... the line between production and development can get a little... blurry. Thanks for the information though, folks. Anathae
mpg321 echoes on HD access
Hallo, my new notebook with DualCore, lots of RAM and SATA harddisk echos on harddisc access like opening thunderbird or firefox with mpg321. I've tried another MP3 Player and changed the BIOS setting from ENHANCED to COMPATIBILITY in the SATA feature, but nothing works really. Thanks in advance Volker Wolfram
Re: bgpd fails to install ipv6 routes in kernel routing table
Claudio Jeker wrote: On Mon, Feb 09, 2009 at 11:43:10AM +0100, Claudio Jeker wrote: On Mon, Feb 09, 2009 at 02:22:08AM -0800, patrick keshishian wrote: On Mon, Feb 9, 2009 at 12:53 AM, Claudio Jeker cje...@diehard.n-r-g.com wrote: On a hunch, I tried a 64bit and a 32 bit machine with 1 prefix each. The 32bit machine adds routes to the kernel without complaint. The 64bit machine complained with send_rtmsg Arrg. IPv6 is once again broken by design. For some ridiculous reason struct sockaddr_in6's size is 28 bytes. So IPv6 fucks up alignment on 64 bit archs. All hail link local addressing and all the crappy workarounds needed for it. Maybe it is too late for me to be thinking about this ... but could you explain the diff below? Unless I'm missing something obvious, it looks like it changes behavior for non-64bit archs as well. Hmm. I think your right. I think a different approach would be better. Will cook up something later today. I think this is better. Just compile tested and no real time to test until later today. Hi Claudio Tested on i386 and amd64 test bgp sessions ok Tested on amd64 production w/2 x ipv4 feeds and 1 x ipv6. Full ipv6 table is installed in the kernel. daemon log shows Feb 10 09:06:14 gw-nextgen bgpd[8598]: neighbor 2001:470:17:7f::1 (HurricaneHK): state change Connect - OpenSent, reason: Connection opened Feb 10 09:06:14 gw-nextgen bgpd[8598]: neighbor 2001:470:17:7f::1 (HurricaneHK): state change OpenSent - OpenConfirm, reason: OPEN message received Feb 10 09:06:14 gw-nextgen bgpd[8598]: neighbor 2001:470:17:7f::1 (HurricaneHK): state change OpenConfirm - Established, reason: KEEPALIVE message received Feb 10 09:06:18 gw-nextgen bgpd[15752]: nexthop 2001:470:17:7f::1 now valid: directly connected No errors.
upgrading packages and ports, ugh
A few months ago I installed amavisd-new by ports. I am now upgrading my system to the latest snapshot (060209). The pkg_add command upgraded many of my packages but left me with packages not upgraded due to them being only available in the ports tree. This seems to me to be a typical scenario so I figured there must be a standard way of dealing with this. Do I actually have to dig and discover what those packages are and then manually re-install them? Probably not but I couldn't find any other way. What I did was take a line from the output of the pkg_add upgrade and apply some tools (grep, sed, cut) to arrive at a file containing one package name per line. Now instead of manually re-installing them can someone help me to programatically upgrade these things? The file contains: expiretable-0.6 freeze-2.5 p5-Archive-Tar-1.38 p5-Compress-Raw-Zlib-2.011 p5-Compress-Zlib-2.011 p5-IO-Compress-Base-2.011 p5-IO-Compress-Zlib-2.011 p5-IO-INET6-2.01p0 p5-IO-Zlib-1.08 postfix-2.5.3-sasl2-mysql unace-1.2bp0 unarj-2.43 unrar-3.81 Ignoring postfix for now (built a certain flavour) the others should be able to be simply upgraded using a script. I'm having trouble turning 'p5-Archive-Tar-1.38' into 'p5-Archive-Tar'. I guess that's what this post boils down to. How to remove the last dash and everything after it. -- jm
Re: upgrading packages and ports, ugh
Hi Juan, Juan Miscaro wrote on Mon, Feb 09, 2009 at 08:38:01PM -0500: A few months ago I installed amavisd-new by ports. I am now upgrading my system to the latest snapshot (060209). The pkg_add command upgraded many of my packages but left me with packages not upgraded due to them being only available in the ports tree. This seems to me to be a typical scenario so I figured there must be a standard way of dealing with this. Do I actually have to dig and discover what those packages are and then manually re-install them? Yes; the reasons for those packages not being upgraded vary, though. Probably not but I couldn't find any other way. What I did was take a line from the output of the pkg_add upgrade and apply some tools (grep, sed, cut) to arrive at a file containing one package name per line. Now instead of manually re-installing them can someone help me to programatically upgrade these things? No, you must do that by hand. Sometimes, things do change, and must be dealt with. Not everything is automatic. The file contains: expiretable-0.6 Obsolete, see pfctl(8), search for -T expire, or see Peter's Book of PF, page 71. p5-Archive-Tar-1.38 p5-Compress-Raw-Zlib-2.011 p5-Compress-Zlib-2.011 p5-IO-Compress-Base-2.011 p5-IO-Compress-Zlib-2.011 p5-IO-Zlib-1.08 http://www.openbsd.org/faq/current.html#20080929 Just remove those. When following -current, also follow the -current FAQ. unace-1.2bp0 unarj-2.43 unrar-3.81 Due to nasty licences, you must build those from source. p5-IO-INET6-2.01p0 freeze-2.5 Don't know those two, sorry. Hope this helps, Ingo
Re: upgrading packages and ports, ugh
On Tue, Feb 10, 2009 at 03:02:28AM +0100, Ingo Schwarze wrote: Hi Juan, Juan Miscaro wrote on Mon, Feb 09, 2009 at 08:38:01PM -0500: [...] p5-IO-INET6-2.01p0 freeze-2.5 Don't know those two, sorry. net/p5-IO-INET6 was replaced by net/p5-IO-Socket-INET6 around the beginning of the year because the upstream CPAN package was renamed. If you already got p5-IO-Socket-INET6 installed while upgrading you may just remove the old p5-IO-INET6 package if it isn't needed anymore. Regards, Markus
usb lan adapter - ADMtek USB To LAN Converter
hi, does anyone knows any of this adapter ? aue0 is created ok, but I can't ping at all. # cat hostname.aue0 inet 10.1.2.30 255.255.255.0 NONE and tcpdump gets something, but can't punt any info on the wire. (was what I could figure out) if anyone can help, TIA matheus # dmesg OpenBSD 4.4 (GENERIC) #1021: Tue Aug 12 17:16:55 MDT 2008 dera...@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: Transmeta(tm) Crusoe(tm) Processor TM5700 (GenuineTMx86 586-class) 799 MHz cpu0: FPU,V86,DE,PSE,TSC,MSR,CX8,SEP,CMOV,SER,MMX real mem = 116924416 (111MB) avail mem = 104345600 (99MB) mainbus0 at root bios0 at mainbus0: AT/286+ BIOS, date 12/10/04, BIOS32 rev. 0 @ 0xfa260, SMBIOS rev. 2.3 @ 0xf0800 (32 entries) bios0: vendor Phoenix Technologies, LTD version 786R1 v1.07 date 12/10/2004 bios0: Hewlett-Packard hp t5000 series acpi0 at bios0: rev 0 acpi0: tables DSDT FACP acpi0: wakeup devices PCI0(S5) LAN0(S5) USB0(S4) USB1(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpiprt0 at acpi0: bus 0 (PCI0) acpicpu0 at acpi0: C3, C2 acpibtn0 at acpi0: PWRB bios0: ROM list: 0xc/0x9000 0xcc000/0x4000 0xd/0xa000 cpu0 at mainbus0 pci0 at mainbus0 bus 0: configuration mode 1 (bios) pchb0 at pci0 dev 0 function 0 Transmeta LongRun Northbridge rev 0x04 Transmeta Mem1 rev 0x00 at pci0 dev 0 function 1 not configured Transmeta Mem2 rev 0x00 at pci0 dev 0 function 2 not configured vendor Transmeta, unknown product 0x0399 (class memory subclass RAM, rev 0x00) at pci0 dev 0 function 3 not configured uhci0 at pci0 dev 9 function 0 VIA VT83C572 USB rev 0x61: irq 11 uhci1 at pci0 dev 9 function 1 VIA VT83C572 USB rev 0x61: irq 15 ehci0 at pci0 dev 9 function 2 VIA VT6202 USB rev 0x63: irq 5 usb0 at ehci0: USB revision 2.0 uhub0 at usb0 VIA EHCI root hub rev 2.00/1.00 addr 1 pciide0 at pci0 dev 11 function 0 CMD Technology SiI3114 SATA rev 0x02: DMA pciide0: using irq 15 for native-PCI interrupt pciide0: port 0: device present, speed: 1.5Gb/s wd0 at pciide0 channel 0 drive 0: SAMSUNG SV2044D wd0: 16-sector PIO, LBA, 19464MB, 39862368 sectors wd0(pciide0:0:0): using BIOS timings, Ultra-DMA mode 4 vga1 at pci0 dev 13 function 0 ATI Radeon VE QY rev 0x00 wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation) wsdisplay0: screen 1-5 added (80x25, vt100 emulation) drm at vga1 unsupported pcib0 at pci0 dev 17 function 0 VIA VT8231 ISA rev 0x10 pciide1 at pci0 dev 17 function 1 VIA VT82C571 IDE rev 0x06: ATA100, channel 0 configured to compatibility, channel 1 configured to compatibility pciide1: channel 0 disabled (no drives) pciide1: channel 1 ignored (disabled) viaenv0 at pci0 dev 17 function 4 VIA VT8231 PMG rev 0x10: failed to map PM I/O space auvia0 at pci0 dev 17 function 5 VIA VT82C686 AC97 rev 0x40: irq 5 ac97: codec id 0x56494161 (VIA Technologies VT1612A) ac97: codec features headphone, 18 bit DAC, 18 bit ADC, KS Waves 3D audio0 at auvia0 vr0 at pci0 dev 18 function 0 VIA RhineII-2 rev 0x51: irq 11, address 00:13:21:ee:26:00 ukphy0 at vr0 phy 1: Generic IEEE 802.3u media interface, rev. 10: OUI 0x004063, model 0x0032 usb1 at uhci0: USB revision 1.0 uhub1 at usb1 VIA UHCI root hub rev 1.00/1.00 addr 1 usb2 at uhci1: USB revision 1.0 uhub2 at usb2 VIA UHCI root hub rev 1.00/1.00 addr 1 isa0 at pcib0 isadma0 at isa0 pckbc0 at isa0 port 0x60/5 pckbd0 at pckbc0 (kbd slot) pckbc0: using irq 1 for kbd slot wskbd0 at pckbd0: console keyboard, using wsdisplay0 pcppi0 at isa0 port 0x61 midi0 at pcppi0: PC speaker spkr0 at pcppi0 npx0 at isa0 port 0xf0/16: reported by CPUID; using exception 16 biomask fdfd netmask fdfd ttymask aue0 at uhub2 port 2 ADMtek USB To LAN Converter rev 1.10/2.01 addr 2 aue0: address 00:60:6e:00:18:c0 ukphy1 at aue0 phy 1: Generic IEEE 802.3u media interface, rev. 1: OUI 0x000749, model 0x0001 uhidev0 at uhub1 port 1 configuration 1 interface 0 GM-TEK USB Composite Device rev 1.01/0.01 addr 2 uhidev0: iclass 3/1 ukbd0 at uhidev0: 8 modifier keys, 6 key codes wskbd1 at ukbd0 mux 1 wskbd1: connecting to wsdisplay0 uhidev1 at uhub1 port 1 configuration 1 interface 1 GM-TEK USB Composite Device rev 1.01/0.01 addr 2 uhidev1: iclass 3/1 ums0 at uhidev1: 3 buttons, Z dir wsmouse0 at ums0 mux 0 softraid0 at root root on wd0a swap on wd0b dump on wd0b -- We will call you cygnus, The God of balance you shall be
Re: Is it possible to increase wscale multiplier?
Grepping through a few log files, the userland program read 44,751,896 bytes with a single syscall. The default recv buf size of 65536 doesn't get the job done for this application. OK, I'll take the bite. The following scenario might apply to your userland app: If your userland program has a very big buffer, it probably takes a while to process it once the read(2) call succeeds. In the meantime new data arrives (which might be dropped on the floor if the sending side does not back off quickly enough because of lack of acks): the kernel buffer fills up to the maximum receive size, and the userland buffer is not available because the new read call has not been issued yet. Again: the sending side is a black box, I cannot change it. The data is generated in real time. I know what real time means. The sending side does not have enough buffer space, therefore it cannot just buffer the data until the BSD box gets around to servicing the Ethernet. So: (1) BSD box's Ethernet driver must run often enough, storing incoming packets in recv buffer and sending acks. (2) userland process must read data often enough so that recv buffer doesn't fill up. Unix makes no guarantee about how often a userland process runs. So the recv buffer must be big enough for the observed worst case plus a safety margin. Setting the recv buffer to 100MB has, so far, been enough. It might be possible to change the userland process so that it could get by with a smaller recv buffer. But the current method works, so I am looking at the remaining problem, which is that sometimes the Ethernet driver gets locked out for too long. I found that resetting a firewire bus would reliably lock out the Ethernet driver causing data loss. It turns out that the culprit is a few lines of text output to the console with printf(9). To isolate the effects of printf(9) from the firewire bus reset, I picked a trivial system call (chown(2)) and added some printf(9) calls. This did not interfere with Ethernet. So printf(9) interferes with Ethernet when called from the firewire driver, but not when called from a vanilla system call. We think there is a locking problem, the firewire device driver expert is working on fixing the locking. Until then, my workaround is to change the printf calls to log(9) calls. There are other events that interfere with the Ethernet driver, but they only happen rarely, so thus far I haven't been able to identify them. If you expect us to help to solving a problem provide details, and not conclusions. I have not asked for help on this list. Again, I haven't tried running this app on OpenBSD yet, I'm running it on FreeBSD. All I did was ask How high is too high?, wondering if OpenBSD had some limit on recv buffer size.
Re: usr.sbin/wake removal
On Mon, 9 Feb 2009 15:14:48 -0500 Ted Unangst ted.unan...@gmail.com wrote: On Mon, Feb 9, 2009 at 12:31 AM, Thomas Pfaff tpf...@tp76.info wrote: I think this could use some explaining for those of us that are not intimately involved in development or have been around here for that long. Keeping it small and simple by saying no to adding one file at 7.2K? I'd really like to know the rationale on this one. I'm kinda amazed at the hoopla over this. Yes, a lot of hoopla; patches flying around, undeadly.org coverage, and then zap for no apparent reason. Not that it matters a great deal, but it does make one raise an eyebrow or three. Last week a wake on lan utility was like the only possible feature not being requested, you didn't even know you wanted it, and now a week later it's like people can't live without it. Yeah, it's handy, but if you survived 10 years without it, I think you can get by a little longer. net/wol has been working for me just fine, so I'm in no need of another utility (although I do like wake better). Thanks. Thomas
