Re: Limit on Alias
* Adam M. Dutko dutko.a...@gmail.com [2011-01-22 19:09]: give it up. you obviously have no idea what you're talking about. an ifaddr is tiny. So what is the base size of one? you can do your homework yourself, really. it is tiny. Can you elaborate how it grows over time based on various levels of traffic? yeah, easily. not at all. -- Henning Brauer, h...@bsws.de, henn...@openbsd.org BS Web Services, http://bsws.de Full-Service ISP - Secure Hosting, Mail and DNS Services Dedicated Servers, Rootservers, Application Hosting
Re: Limit on Alias
* Adam M. Dutko dutko.a...@gmail.com [2011-01-22 00:00]: Hahaha. I don't understand the humor. I've had over 300k addresses on a single interface in a test environment before. Very cool, so it was a test environment. Did you roll it to production? How well did it work? Like Henning said, the limit is memory. I imagine memory would be a big factor. I guess I should have added that as a qualifier but in general unless you have gobs of RAM more than a few hundred in production might be an issue. give it up. you obviously have no idea what you're talking about. an ifaddr is tiny. -- Henning Brauer, h...@bsws.de, henn...@openbsd.org BS Web Services, http://bsws.de Full-Service ISP - Secure Hosting, Mail and DNS Services Dedicated Servers, Rootservers, Application Hosting
Re: Limit on Alias
give it up. you obviously have no idea what you're talking about. an ifaddr is tiny. So what is the base size of one? Can you elaborate how it grows over time based on various levels of traffic?
Re: Limit on Alias
Adam M. Dutko dutko.a...@gmail.com writes: give it up. you obviously have no idea what you're talking about. an ifaddr is tiny. So what is the base size of one? Can you elaborate how it grows over time based on various levels of traffic? grep in your /usr/src/sys for '.h files containing struct ifaddr, see what it turns up and work from there. -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/ Remember to set the evil bit on all malicious network traffic delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
Limit on Alias
What it's the limit of number alias that a single ethernet interface can support?
Re: Limit on Alias
* Orestes Leal R. l...@cubacatering.avianet.cu [2011-01-21 15:50]: What it's the limit of number alias that a single ethernet interface can support? memory. -- Henning Brauer, h...@bsws.de, henn...@openbsd.org BS Web Services, http://bsws.de Full-Service ISP - Secure Hosting, Mail and DNS Services Dedicated Servers, Rootservers, Application Hosting
Re: Limit on Alias
What it's the limit of number alias that a single ethernet interface can support? I believe 254?
Re: Limit on Alias
On 2011 Jan 21 (Fri) at 09:51:52 -0500 (-0500), Adam M. Dutko wrote: : What it's the limit of number alias that a single ethernet interface can : support? : :I believe 254? : Hahaha. I've had over 300k addresses on a single interface in a test environment before. Like Henning said, the limit is memory. -- MacDonald has the gift on compressing the largest amount of words into the smallest amount of thoughts. -- Winston Churchill
Re: Limit on Alias
On 2011-01-21 18.42, Peter Hessler wrote: On 2011 Jan 21 (Fri) at 09:51:52 -0500 (-0500), Adam M. Dutko wrote: : What it's the limit of number alias that a single ethernet interface can : support? : :I believe 254? Hahaha. I've had over 300k addresses on a single interface in a test environment before. Like Henning said, the limit is memory. What about performance? Is there any noticable impact with let's say a couple of hundred alias addresses? /B -- internetlabbet.se / work: +46 8 551 124 80 / Words must Benny Lofgren/ mobile: +46 70 718 11 90 / be weighed, / fax:+46 8 551 124 89/not counted. /email: benny -at- internetlabbet.se
Re: Limit on Alias
* Benny Lofgren bl-li...@lofgren.biz [2011-01-21 19:33]: On 2011-01-21 18.42, Peter Hessler wrote: On 2011 Jan 21 (Fri) at 09:51:52 -0500 (-0500), Adam M. Dutko wrote: : What it's the limit of number alias that a single ethernet interface can : support? : :I believe 254? Hahaha. I've had over 300k addresses on a single interface in a test environment before. Like Henning said, the limit is memory. What about performance? Is there any noticable impact with let's say a couple of hundred alias addresses? there is an impact, ifaddrs are kept in a simple linked list. in -current there is also an RB tree (which makes the impact much much much smaller) but it isn't used everywhere yet. -- Henning Brauer, h...@bsws.de, henn...@openbsd.org BS Web Services, http://bsws.de Full-Service ISP - Secure Hosting, Mail and DNS Services Dedicated Servers, Rootservers, Application Hosting
Re: Limit on Alias
Hahaha. I don't understand the humor. I've had over 300k addresses on a single interface in a test environment before. Very cool, so it was a test environment. Did you roll it to production? How well did it work? Like Henning said, the limit is memory. I imagine memory would be a big factor. I guess I should have added that as a qualifier but in general unless you have gobs of RAM more than a few hundred in production might be an issue.
