gmane.org.user-groups.ale

bonding two broadband connections

Geoffrey — 2008-10-07T19:56:06

I'm looking for a firewall product (software or hardware) of reasonable price that will take two broadband connections and bond them together. Does anyone have anyone experience with such a product? I know smoothwall does, but it's $2k, and that's well outside the price range I'm considering.

Re: Considerations Regarding Dual Power Supplies

Greg Freemyer — 2008-10-07T19:40:45

We use to worry about them a lot when I built avionics trainers (late 80s). I think the problem was the low-voltage single ended data signals we had running around. I suspect it can still be a problem, but technologies like SATA/USB/firewire use differential signals which let the controller and the drive be at different ground potentials. (Or so I understand.) Greg

Re: user limits

Jim Kinney — 2008-10-07T19:35:54

This is weird. Pam can block users from chewing up too many resources but where does it output it's signaling? I can set a low limit on noprocs and login and get a "Resource Temporarily Unavailable" message on the console. But that's useless if it's a mouse click on a launcher icon... On Tue, 2008-10-07 at 19:23 +0000, hbbs-Wuw85uim5zDR7s880joybQ< at >public.gmane.org wrote:

Re: user limits

Warren Myers — 2008-10-07T19:34:34

side note: Itankum may be somewhat iffy, but it's the exclusive platform for HP-UX going forward. HP has dropped the PA-RISC architecture in favor of IA64. WMM On Tue, Oct 7, 2008 at 3:23 PM, public.gmane.org> wrote:

Re: user limits

Tim Meanor — 2008-10-07T19:25:31

I believe that when a user hits a soft limit, the same thing happens as when they hit a hard limit - in the case of a memory soft limit, a segfault. The difference is that a soft limit can be changed by a non-root user on-the-fly, up to the hard limit. -Tim On Tue, Oct 7, 2008 at 2:04 PM, Jim Kinney public.gmane.org> wrote:

Re: user limits

2008-10-07T19:23:46

For what it's worth, this behavior is native to OpenVMS...it's just too bad that the OS is limited to Itanium and Itanium's future is...iffy. - Jeff -------------- Original message ---------------------- From: Jim Kinney public.gmane.org>

Re: user limits

Jim Kinney — 2008-10-07T18:04:57

xmessage is the way to do it. Now to find a daemon that can do the deed and deliver an intelligent message.... or write one which is looking like the real solution. Need to find _where_ the soft limit signal is written (/var/log/message? STDOUT?) On Tue, 2008-10-07 at 13:55 -0400, James Sumners wrote:

Re: user limits

James Sumners — 2008-10-07T17:55:13

Launch everything via a script that checks the limit and then issues an xmessage[1]? [1] -- http://www.manpagez.com/man/1/xmessage/ On Tue, Oct 7, 2008 at 1:45 PM, Jim Kinney public.gmane.org> wrote:

user limits

Jim Kinney — 2008-10-07T17:45:24

When a system uses pam, the file (typically) /etc/security/limits.conf can control limits for things like resident set size, stack size, number processes, login count, etc. What happens when a user hits a soft limit? A hard limit means the operation is not completed and/or fails. I'm looking for a way to notify users that they are pushing their luck opening another (eg - firefox) application as they are running out of ram. Jim Kinney

Re: RSS + TTL

Chris Fowler — 2008-10-07T14:38:19

_______________________________________________ Ale mailing list Ale-S6NtOCTnm14< at >public.gmane.org http://mail.ale.org/mailman/listinfo/ale

Re: RSS + TTL

Jim Popovitch — 2008-10-07T13:47:51

I've never seen a ttl stanza in an rss feed, but the following should work for 2 minute updates: SAM AlarmCenter Organization: DEMO http://demo.opsdc.com/ hourly 30 1901-01-01T00:00+00:00 -Jim P.

Re: Documents and VCS (was Re: Anyone looking for a new gig?)

Jeff Lightner — 2008-10-07T13:24:42

I'm not pushing use of Sharepoint - just responding to an earlier comment about Sharepoint. -----Original Message----- From: ale-bounces-S6NtOCTnm14< at >public.gmane.org [mailto:ale-bounces-S6NtOCTnm14< at >public.gmane.org] On Behalf Of George L. Allen Sent: Monday, October 06, 2008 6:03 PM To: ale-S6NtOCTnm14< at >public.gmane.org Subject: Re: [ale] Documents and VCS (was Re: Anyone looking for a new gig?) On Mon, Oct 06, 2008 at 10:55:45AM -0400, Jeff Lightner wrote: out, save But I think it works by storing *each* version rather than diff/merge-able deltas between versions. (Nifty vs. Useful) I think mediawiki is probably the best-of, probably with a WYSIWYG plugin so the users don't have to wrap their brains around markup. _______________________________________________ Ale mailing list Ale-S6NtOCTnm14< at >public.gmane.org http://mail.ale.org/mailman/listinfo/ale ---------------------------------- CONFIDENTIALITY NOTICE: This e-mail may contain privileged or confidential information and is for the sole use of the intended recipient(s). If you are not the intended recipient, any disclosure, copying, distribution, or use of the contents of this information is prohibited and may be unlawful. If you have received this electronic transmission in error, please reply immediately to the sender that you have received the message in error, and delete it. Thank you. ----------------------------------

RSS + TTL

Chris Fowler — 2008-10-07T13:18:27

Anyone here playing around with RSS? I decided to do a little research today to see how I can use RSS in our system. Most of the items I want the client to see are time sensitive and I need the client to read the XML file on the server at least once per 5 minutes. I've added the option the feed and it does not seem FireFox 3.0 respects it. I'm tailing apaches access log and I see no attempt to grab the file every 5 minutes. Possibly I've did something wrong? Here is the top of the rss feed file SAM AlarmCenter Organization: DEMO http://demo.opsdc.com/ 5 support-C5vNwMtvQp34lQ16zRZlbA< at >public.gmane.org Maybe FF has a minimum?

Re: Considerations Regarding Dual Power Supplies

Jim Lynch — 2008-10-07T11:07:17

I'm going to hazard a guess that most power supplies are well filtered and regulated so even if ground loops exist, they are so far down in the noise, they can be ignored, unless the power distribution systems have really crappy ground systems. In the case of other electronic equipment like audio amplifiers and transmitters, a ground loop can sometimes be a real challenge. I've been around computer installations for many years now and can't ever recall ground loops being a problem or even being discussed. I agree with the others about redundancy being probably a more important consideration. Jim.

Re: Documents and VCS (was Re: Anyone looking for a new gig?)

George L. Allen — 2008-10-06T22:03:17

But I think it works by storing *each* version rather than diff/merge-able deltas between versions. (Nifty vs. Useful) I think mediawiki is probably the best-of, probably with a WYSIWYG plugin so the users don't have to wrap their brains around markup.

Re: Considerations Regarding Dual Power Supplies

Jeff Lightner — 2008-10-06T21:03:45

I wasn't saying he should worry about it but simply that I'd seen a presentation once that talked a lot about grounding. It talked some about how people would run one long ground to one location then a short ground to someplace else and be surprised when a short occurred that caused all the power to suddenly go through that short ground and essentially cause it to become a live circuit. The presenter talked about things such as shunts etc... that allow you to connect ground sources so that the shortest path is still to ground to prevent this. Having seen a whole data center go off line with cascade failures from one panel to another until all of them were fried I know that power considerations aren't trivial. I certainly wouldn't present myself as an expert on power because I'm not. I was just trying to answer the idea of grounding and redundancy in the OP's post. When I deal with multiple PDUs I seldom ask myself "where is the ground for this" as I'm relying on the electrical folks that create the circuits I'm plugging the PDUs into to have done their job properly. -----Original Message----- From: ale-bounces-S6NtOCTnm14< at >public.gmane.org [mailto:ale-bounces-S6NtOCTnm14< at >public.gmane.org] On Behalf Of Greg Freemyer Sent: Monday, October 06, 2008 4:13 PM To: ale-S6NtOCTnm14< at >public.gmane.org Subject: Re: [ale] Considerations Regarding Dual Power Supplies Jeff, I don't know why ground loops are not typically a big issue with servers and dual power supplies, but I have RARELY seen anyone worry about it. The exception is, I have seen a 30 foot shaft sunk into the ground near a computer room and a ground line from that directly connected to a ground distribution panel, which in turn was connected out to the servers. Greg On Mon, Oct 6, 2008 at 3:33 PM, public.gmane.org> wrote: i.e, a closed circuit of what should be ground in which EMI (especially 60Hz hum and harmonics) can be induced. machine with two power supplies to two different PDUs in a server rack? Or, is it preferable to connect the two power cables to as close to the same point as possible (i.e., two electrically adjacent outlets in the same PDU)? Aren't the ground pins in each of the two power supplies going to be connected together inside the machine? If so, it seems as though having the two grounds form a big loop might bring a lot of hash into the machine through the ground lines.

Re: Considerations Regarding Dual Power Supplies

Greg Freemyer — 2008-10-06T20:12:46

Jeff, I don't know why ground loops are not typically a big issue with servers and dual power supplies, but I have RARELY seen anyone worry about it. The exception is, I have seen a 30 foot shaft sunk into the ground near a computer room and a ground line from that directly connected to a ground distribution panel, which in turn was connected out to the servers. Greg On Mon, Oct 6, 2008 at 3:33 PM, public.gmane.org> wrote:

Re: Considerations Regarding Dual Power Supplies

Jim Kinney — 2008-10-06T20:06:36

Ditto. Just be sure that the ground on each PS at the server makes a SOLID connection to a common point on the chassis and that the ground in the separate power feeds is a common one. On Mon, 2008-10-06 at 15:53 -0400, Jeff Lightner wrote:

Re: Considerations Regarding Dual Power Supplies

Jeff Lightner — 2008-10-06T19:53:22

You should certainly go to two separate PDUs with separate power circuits for redundancy. We do that with all of our servers specifically to insure loss of one PDU or circuit doesn't take servers down. I did see a presentation once on power that talked about the important thing being that everything had a common ground. It seems if both circuits go to the same ground (and it is the shortest path) you don't have to worry about separate PDUs. If the PDUs are daisy chained together they're all on the same circuit anyway so there's no benefit to spreading the load between them. I've seen some racks that do things that way (often due to poor planning). -----Original Message----- From: ale-bounces-S6NtOCTnm14< at >public.gmane.org [mailto:ale-bounces-S6NtOCTnm14< at >public.gmane.org] On Behalf Of hbbs-Wuw85uim5zDR7s880joybQ< at >public.gmane.org Sent: Monday, October 06, 2008 3:33 PM To: ale-S6NtOCTnm14< at >public.gmane.org Subject: [ale] Considerations Regarding Dual Power Supplies In audio-land, you have to be mindful of not creating ground loops, i.e, a closed circuit of what should be ground in which EMI (especially 60Hz hum and harmonics) can be induced. Knowing that, does it cause problems to run power cables from a machine with two power supplies to two different PDUs in a server rack? Or, is it preferable to connect the two power cables to as close to the same point as possible (i.e., two electrically adjacent outlets in the same PDU)? Aren't the ground pins in each of the two power supplies going to be connected together inside the machine? If so, it seems as though having the two grounds form a big loop might bring a lot of hash into the machine through the ground lines. - Jeff _______________________________________________ Ale mailing list Ale-S6NtOCTnm14< at >public.gmane.org http://mail.ale.org/mailman/listinfo/ale ---------------------------------- CONFIDENTIALITY NOTICE: This e-mail may contain privileged or confidential information and is for the sole use of the intended recipient(s). If you are not the intended recipient, any disclosure, copying, distribution, or use of the contents of this information is prohibited and may be unlawful. If you have received this electronic transmission in error, please reply immediately to the sender that you have received the message in error, and delete it. Thank you. ----------------------------------

Re: Considerations Regarding Dual Power Supplies

Stephen Benjamin — 2008-10-06T19:45:21

_______________________________________________ Ale mailing list Ale-S6NtOCTnm14< at >public.gmane.org http://mail.ale.org/mailman/listinfo/ale

Considerations Regarding Dual Power Supplies

2008-10-06T19:33:14

In audio-land, you have to be mindful of not creating ground loops, i.e, a closed circuit of what should be ground in which EMI (especially 60Hz hum and harmonics) can be induced. Knowing that, does it cause problems to run power cables from a machine with two power supplies to two different PDUs in a server rack? Or, is it preferable to connect the two power cables to as close to the same point as possible (i.e., two electrically adjacent outlets in the same PDU)? Aren't the ground pins in each of the two power supplies going to be connected together inside the machine? If so, it seems as though having the two grounds form a big loop might bring a lot of hash into the machine through the ground lines. - Jeff