[gentoo-user] Re: [Fwd: Re: Gentoo Rules]
On Fri, 14 Dec 2007 21:07:53 -0500, Randy Barlow wrote: 7v5w7go9ub0o wrote: My concerns with this, other than my abilities, are: 1. Showing proper respect to the guy who pioneered the effort to date, and who may simply be out of town. (This disrespect would be alleviated if there was an official policy encouraging volunteer ebuilds.) It's not disrespectful, IMO, to do something that you don't see getting done. Especially since it's less work for another guy. I wouldn't worry about that point. As a developer I agree with that point. It's always better to get bug reports for version bumps or problems that have patches attached to them, or even a simple note saying that you copied the ebuild to the new version and things work fine. This can happen. I've submitted ebuilds for backuppc-3.0.0, and so have many other people. In fact, the bug for it has several ebuilds that have been submitted but haven't made it into the official tree. I think that particular bug report might not be getting attention from the right people or something. That doesn't mean it isn't worth doing though, because people can still use the ebuild from the bug report. Ideally, a dev would see that, check it out for correctness, and add it to ~arch. I guess you are talking about https://bugs.gentoo.org/show_bug.cgi? id=141018 ? It's assigned to maintainer-needed (ie, it is in the tree, but currently no developer is maintaining it). The original maintainer recently retired, so it is now in some sort of limbo. In this case the fallback would be the backup herd (who are listed on the bug), but I know that these folks are understaffed. As you can tell from this we are always looking for more developers. Does anybody know how to call attention to a bug report that doesn't seem to have any devs paying attention to it? I think BackupPC is a fine product, and would like to see it in the tree for others to use. I'm using my own ebuild successfully, as are many of the fine folks who have contributed on that bug report. I'd just like my and others' efforts to be something that benefits more of the Gentoo community :) A possible solution would be for you (or someone) to become a proxy maintainer, meaning that you'd get the bug reports and provide new ebuilds, and a developer (most likely someone from the backup herd) would review it and put it in the tree. Kind regards, Hans -- [EMAIL PROTECTED] mailing list
[gentoo-user] Re: Gentoo Rules
On Fri, 14 Dec 2007 11:56:41 -0800, Grant wrote: Lately I've been shopping around for other distros as well as looking at *BSD. Gentoo development seems to have slowed way down and I like things being improved as quickly as possible. FreeBSD is supposed to be the closest relation, but even that won't do. I don't think there is anything as satisfying as Gentoo out there. The concept is second to none, the execution of that concept is fantastic, but it needs to keep moving forward. What is the next step? Or should we keep treading water? - Grant I love gentoo and can't settle for anything else. What can I do to make sure development doesn't stop? Let me in on that. What can I do too? There are plenty of things that can be done, depending on what kind of skills you bring with you. And please note that those skills need not be technical in order to help out. Just some things off the top of my head: * participate in the community (e.g. here or in the forums) to help others with Gentoo things * participate on bugs.gentoo.org by adding relevant comments to bugs, trying to fix bugs, providing new ebuilds or patches (and bugday is a good way to get started with that: http://bugday.gentoo.org/) * help out the documentation teams to maintain the current information or create new stuff and possible translate it * help out with Gentoo artwork * help out with the organization of Gentoo stuff such as events and PR * becoming a developer: http://www.gentoo.org/proj/en/devrel/staffing- needs/ * that one thing that you can do really well but that I forgot to list here Feel free to drop me an email off-list if you'd like to discuss what you can do for Gentoo. Kind regards, Hans -- [EMAIL PROTECTED] mailing list
[gentoo-user] Re: Gentoo Rules
On Sat, 15 Dec 2007 01:05:08 +0100, b.n. wrote: Florian Philipp ha scritto: Other things to improve? A better documentation on USE-flags. In my opinion every maintainer should provide as much information as possible on what exactly a USE-flag changes. At the moment it's the administrator's responsibility to find this out. Not really a good idea on production systems if you ask me ... +1 m. Good news then as a scheme for this has been proposed and partially implemented: http://blog.cardoe.com/archives/2007/11/19/use-flag-metadata/ It was decided in the last council meeting to keep this scheme: http:// www.gentoo.org/proj/en/council/meeting-logs/20071213-summary.txt This only provides the information, it may take some time before user- facing tools (such as euse) expose this information, and obviously developers need to add the information. Kind regards, Hans -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
On Fri, 2007-12-14 at 18:13 -0600, Dale wrote: Florian Philipp wrote: Okay, here it goes: I think we could need a better support for binary packages. There was a thread in here a few months ago about how to offer binary packages for customers. As far as I remember the problem was (and still is) that there is no easy way to check the packages for corruption (trojans, stuff like that). I know some things are only available as a binary but Gentoo is about compiling your own packages. Binaries are for Redhat, Mandrake and such. I moved away from that for good reason. As far as I know Gentoo is all about choice. Shouldn't the user have the choice to use binary packages? Maybe his/her laptop doesn't stand the thermal output of its CPU when emerging or maybe he/she's the administrator of a large company's network, trying to move every computer system to Gentoo. If you want Gentoo to grow and prosper, you must accept that there are people who don't want or are not able to compile every piece of software but still want to use Gentoo. Just my two cents. -Florian Philipp signature.asc Description: PGP signature
Re: [gentoo-user] Gentoo Rules
Florian Philipp wrote: On Fri, 2007-12-14 at 18:13 -0600, Dale wrote: Florian Philipp wrote: Okay, here it goes: I think we could need a better support for binary packages. There was a thread in here a few months ago about how to offer binary packages for customers. As far as I remember the problem was (and still is) that there is no easy way to check the packages for corruption (trojans, stuff like that). I know some things are only available as a binary but Gentoo is about compiling your own packages. Binaries are for Redhat, Mandrake and such. I moved away from that for good reason. As far as I know Gentoo is all about choice. Shouldn't the user have the choice to use binary packages? Maybe his/her laptop doesn't stand the thermal output of its CPU when emerging or maybe he/she's the administrator of a large company's network, trying to move every computer system to Gentoo. If you want Gentoo to grow and prosper, you must accept that there are people who don't want or are not able to compile every piece of software but still want to use Gentoo. Just my two cents. -Florian Philipp That is when you compile it on another machine then install it on the laptop. The -K option comes to mind here. I also think that the choice is in what you install as far as programs and the options they have available. Gentoo is Linux from Scratch with a serious package manager. Choice is not about having binaries or not. Also keep in mind that if a binary has something compiled in that you don't want or need, you are stuck with it and its dependencies. If you compile your own, you have choices. This and dependency he** is why I switched from Mandrake, now known as Mandriva or something like that. My two cents. Dale :-) :-) :-)
Re: [gentoo-user] Gentoo Rules
I love gentoo and can't settle for anything else. What can I do to make sure development doesn't stop? Let me in on that. What can I do too? Help out with bugfixing by submitting patches or even just confirming bugs and supplying needed details? Join testing teams? Join the Weekly News team? Is it really so difficult to see where there is something to do? If yes, wouldn't it be the first step to learn about how it all works? ralf -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
On Sat, 15 Dec 2007 03:44:55 -0600, Dale wrote: That is when you compile it on another machine then install it on the laptop. The -K option comes to mind here. Which is what I think the OP was talking about. If you install one of the *-bin packages from portage, you are protected by the checksums in the ebuild digest. But if you create a binary package repository, there is currently no means of applying the same protection. So if you are administering machines at different locations and want to keep a single binary package repository so you only build once (remember, production servers may not have gcc installed), there is no means of checking that the downloaded package has not been tampered with. This protection applies to ebuilds and distfiles but cannot be applied to packages you build yourself. I also think that the choice is in what you install as far as programs and the options they have available. Gentoo is Linux from Scratch with a serious package manager. Choice is not about having binaries or not. Also keep in mind that if a binary has something compiled in that you don't want or need, you are stuck with it and its dependencies. This is not about precompiled packages from a distro. Portage already has the mechanism for build once, install many, it is just lacking some of the safeguards at the install stage that are present for the build stage. -- Neil Bothwick Computers are like Old Testament gods; lots of rules and no mercy. signature.asc Description: PGP signature
Re: [gentoo-user] Gentoo Rules
On Saturday 15 December 2007 03:35:51 Grant wrote: My ideas aren't really important unless they're everyone else's ideas too. What is it exactly you want to achieve by starting these pointless threads? -- Bo Andresen signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Gentoo Rules
Neil Bothwick wrote: On Sat, 15 Dec 2007 03:44:55 -0600, Dale wrote: That is when you compile it on another machine then install it on the laptop. The -K option comes to mind here. Which is what I think the OP was talking about. If you install one of the *-bin packages from portage, you are protected by the checksums in the ebuild digest. But if you create a binary package repository, there is currently no means of applying the same protection. So if you are administering machines at different locations and want to keep a single binary package repository so you only build once (remember, production servers may not have gcc installed), there is no means of checking that the downloaded package has not been tampered with. This protection applies to ebuilds and distfiles but cannot be applied to packages you build yourself. But he was responding to me mentioning Redhat and Mandrake which are binary based. Maybe I took his original point wrong. I also think that the choice is in what you install as far as programs and the options they have available. Gentoo is Linux from Scratch with a serious package manager. Choice is not about having binaries or not. Also keep in mind that if a binary has something compiled in that you don't want or need, you are stuck with it and its dependencies. This is not about precompiled packages from a distro. Portage already has the mechanism for build once, install many, it is just lacking some of the safeguards at the install stage that are present for the build stage. True, but I was comparing to distros that are binary based not that you compile yourself. Again, the Redhat and Mandrake type of thing. I wish I had a laptop sometimes. Then sometimes I'm glad I have my desktop. Dale :-) :-)
Re: [gentoo-user] Gentoo Rules
That is when you compile it on another machine then install it on the laptop. The -K option comes to mind here. Which is what I think the OP was talking about. If you install one of the *-bin packages from portage, you are protected by the checksums in the ebuild digest. But if you create a binary package repository, there is currently no means of applying the same protection. So if you are administering machines at different locations and want to keep a single binary package repository so you only build once (remember, production servers may not have gcc installed), there is no means of checking that the downloaded package has not been tampered with. This protection applies to ebuilds and distfiles but cannot be applied to packages you build yourself. But he was responding to me mentioning Redhat and Mandrake which are binary based. Maybe I took his original point wrong. Exactly :) Neil correctly translated my pseudo-English to what I actually meant. I don't want to make Portage binary based. I just want to make Portage's binary package support more conveniently usable on big networks. I don't think there is any shortage of great ideas here. Can we get into specifics on how projects are born and become successful? So, what would need to happen for one of these projects to take off would be one or more people to be in charge of it and organize it, and they recruit as many people as possible to work on the project along with them? Does that recruitment generally take the form of volunteers finding the project as opposed to the project finding volunteers? Any light to shed on this process for me? - Grant -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: DMZ on an vmware gentoo guest running on winXP host
On Friday 14 December 2007, [EMAIL PROTECTED] wrote: Mick [EMAIL PROTECTED] writes: Would I likely be opening my lan up for some christmas shopping by having a gentoo guest on a WinXP host running as a DMZ machine? It would be pretty barebones with a IPTABLE setup for logging and tagging or whatever I get interested in doing with the traffic. No X server or other frills. A rather simpler solution to do this would be to get hold of hub, connect it to the firewall and watch everything that passes through it. I do have an older hub, but not sure what you mean here. The hub has no network address and of course is not switched so anything going thru it can be filtered with tcpdump. But the router is switched. Not sure how a hub would see the outfacing address. I'd be able to see all the lan machines that were going thru it, but how about the traffic that the firewall is rejecting? Thats what I'm after. Can you elaborate a little? Maybe you mean something different by `hub'. I mean a hardware hub, not a switch and not a router. You need to place it in-line between your router/switch and your modem. Being on the WAN side of your NAT it will 'see' all the packets that go to/from the Internet (unfiltered). On the other side of the router you get the filtered traffic which when compared/contrasted with the WAN side will show you what the router and it's firewall are doing. I hope this is a bit clearer, otherwise please email me if you think this is getting off topic. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Re: Gentoo Rules
Lately I've been shopping around for other distros as well as looking at *BSD. Gentoo development seems to have slowed way down and I like things being improved as quickly as possible. FreeBSD is supposed to be the closest relation, but even that won't do. I don't think there is anything as satisfying as Gentoo out there. The concept is second to none, the execution of that concept is fantastic, but it needs to keep moving forward. What is the next step? Or should we keep treading water? - Grant I love gentoo and can't settle for anything else. What can I do to make sure development doesn't stop? Let me in on that. What can I do too? There are plenty of things that can be done, depending on what kind of skills you bring with you. And please note that those skills need not be technical in order to help out. Just some things off the top of my head: * participate in the community (e.g. here or in the forums) to help others with Gentoo things * participate on bugs.gentoo.org by adding relevant comments to bugs, trying to fix bugs, providing new ebuilds or patches (and bugday is a good way to get started with that: http://bugday.gentoo.org/) * help out the documentation teams to maintain the current information or create new stuff and possible translate it * help out with Gentoo artwork * help out with the organization of Gentoo stuff such as events and PR * becoming a developer: http://www.gentoo.org/proj/en/devrel/staffing- needs/ * that one thing that you can do really well but that I forgot to list here What keeps an army of developers from putting in more time on a cool project? I'll bet there are many Gentoo developers who would rather work on Gentoo than most other things. These developers who pretty much *are* Gentoo aren't able to focus on Gentoo because living costs money and Gentoo doesn't pay. Multiple great ideas have already been suggested in this thread. Is this the first time they've been conceived and shared? Why hasn't work begun on them? Why isn't work completed on them? Because living costs money and Gentoo doesn't pay. I've been in business for 7 years and I'd like to take a shot at designing a system that would pay Gentoo developers to develop. Flame me. I can take it. Long live Gentoo. - Grant -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
On Saturday 15 December 2007 15:05:28 Grant wrote: Neil correctly translated my pseudo-English to what I actually meant. I don't want to make Portage binary based. I just want to make Portage's binary package support more conveniently usable on big networks. Even eclasses in the tree don't have any sort of checksums and they aren't even included in binary packages either... I don't think there is any shortage of great ideas here. Can we get into specifics on how projects are born and become successful? So, what would need to happen for one of these projects to take off would be one or more people to be in charge of it and organize it, and they recruit as many people as possible to work on the project along with them? The real blocker for features that I'd like Gentoo to support is Portage. There is only 1½ people working on it and changing anything in it is hard because Portage is a horrible mess. There's plenty of activity in the tree but new desired features cannot be used in the tree until Portage supports them. It also doesn't make matters better that over the years all sorts of weird hacks (that now have to be supported) have been added to the tree instead of waiting for proper solutions. Most people who are capable of helping to improve Portage just don't want to touch it. Does that recruitment generally take the form of volunteers finding the project as opposed to the project finding volunteers? Any light to shed on this process for me? If there's one thing we definitely don't need it's more clueless people who become developers just because they claim they want to do something. Being stalled is better than major screw ups that hurt everyone and than moving in the wrong direction. -- Bo Andresen signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
Hans de Graaff wrote: A possible solution would be for you (or someone) to become a proxy maintainer, meaning that you'd get the bug reports and provide new ebuilds, and a developer (most likely someone from the backup herd) would review it and put it in the tree. Hi Hans, thanks for the reply! Being a student, I will soon have some time to help out, so this might be a good thing for me to do. How does one become a proxy maintainer? Should I just write [EMAIL PROTECTED] One of the challenging things about the BackupPC ebuild is that the program needs to be configured to work with its own instance of apache (run as user backuppc), and I think none of the ebuild contributors are all too sure of the standard Gentoo way of doing this. I'd be happy to try and make an ebuild that is a good compromise of the ideas listed in the bug. -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] [OT] Network Trouble
On Saturday 15 December 2007, Hilco Wijbenga wrote: On Dec 14, 2007 3:46 PM, Kenneth Prugh [EMAIL PROTECTED] wrote: Are you using the sky2 driver for your NIC by chance? Nope, I've tried a tulip and a 3COM NIC. Pretty run-of-the-mill. :-) Since you seem to have checked that the problem is not related to cables/NICs/modem/router - does the machine work fine otherwise? If you had bad memory/fs corruption during your download when the NIC failed it may be that you need to remerge some parts of your system. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Gentoo Rules
Florian Philipp wrote: Maybe his/her laptop doesn't stand the thermal output of its CPU when emerging or maybe he/she's the administrator of a large company's network, trying to move every computer system to Gentoo. Check out distccd! -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
On Saturday 15 December 2007, Randy Barlow wrote: One of the challenging things about the BackupPC ebuild is that the program needs to be configured to work with its own instance of apache (run as user backuppc), and I think none of the ebuild contributors are all too sure of the standard Gentoo way of doing this. I'd be happy to try and make an ebuild that is a good compromise of the ideas listed in the bug. (Apologies if this has already been mentioned) Another (maybe less intrusive, although slightly less efficient) option is to install the BackupPC_Admin CGI as setuid so that it runs as user backuppc (this is how I run BackupPC-2.1.2-r1). This does not require a different instance of apache, nor it requires to use mod_perl. It does, however, require USE=perlsuid, but this is already supported by gentoo. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
Etaoin Shrdlu wrote: (Apologies if this has already been mentioned) Another (maybe less intrusive, although slightly less efficient) option is to install the BackupPC_Admin CGI as setuid so that it runs as user backuppc (this is how I run BackupPC-2.1.2-r1). This does not require a different instance of apache, nor it requires to use mod_perl. It does, however, require USE=perlsuid, but this is already supported by gentoo. Yes, I am aware of that. The BackupPC ebuild should support either way, as there is a speedup of about 15x (according to the BackupPC author) when running the webserver as user backuppc. There should be a USE variable controlling this. Also, it is wise to not put backuppc on the same instance of apache so that it can run on a non-standard port. The ebuild by default would not install any password control to backuppc, and so the web interface would be open. -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
On Saturday 15 December 2007, Randy Barlow wrote: Yes, I am aware of that. The BackupPC ebuild should support either way, as there is a speedup of about 15x (according to the BackupPC author) when running the webserver as user backuppc. There should be a USE variable controlling this. Also, it is wise to not put backuppc on the same instance of apache so that it can run on a non-standard port. The ebuild by default would not install any password control to backuppc, and so the web interface would be open. Ah ok, I just thought it would be easier, to get things going and catch up with upstream, to release an ebuild that only supports the suid mode of operation, and then, taking the necessary time, improve it in future releases, rather than supporting all the features right from the beginning. But of course this is just my personal opinion. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
Etaoin Shrdlu wrote: Ah ok, I just thought it would be easier, to get things going and catch up with upstream, to release an ebuild that only supports the suid mode of operation, and then, taking the necessary time, improve it in future releases, rather than supporting all the features right from the beginning. But of course this is just my personal opinion. Yeah, that's the kinds of differences of opinion that are in the bug report, which is part of what makes this a more difficult ebuild to write. Things like libraries are really easy because it's just a configure make make install, but here you have a lot of configuration files and differences of opinion. I was thinking that a USE variable could be in order here, to support suid and a separate apache instance. Perhaps the variable could be suid? -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
So, what would need to happen for one of these projects to take off would be one or more people to be in charge of it and organize it, and they recruit as many people as possible to work on the project along with them? The real blocker for features that I'd like Gentoo to support is Portage. There is only 1½ people working on it and changing anything in it is hard because Portage is a horrible mess. There's plenty of activity in the tree but new desired features cannot be used in the tree until Portage supports them. It also doesn't make matters better that over the years all sorts of weird hacks (that now have to be supported) have been added to the tree instead of waiting for proper solutions. Most people who are capable of helping to improve Portage just don't want to touch it. Would you say that portage is the main block in the way of Gentoo's continued progress? Does that recruitment generally take the form of volunteers finding the project as opposed to the project finding volunteers? Any light to shed on this process for me? If there's one thing we definitely don't need it's more clueless people who become developers just because they claim they want to do something. Being stalled is better than major screw ups that hurt everyone and than moving in the wrong direction. I don't have aspirations of becoming a developer if that's what you mean. Thanks though. - Grant -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
The real blocker for features that I'd like Gentoo to support is Portage. There is only 1½ people working on it and changing anything in it is hard because Portage is a horrible mess. There's plenty of activity in the tree but new desired features cannot be used in the tree until Portage supports them. It also doesn't make matters better that over the years all sorts of weird hacks (that now have to be supported) have been added to the tree instead of waiting for proper solutions. Most people who are capable of helping to improve Portage just don't want to touch it. Would you say that portage is the main block in the way of Gentoo's continued progress? Actually I guess that's pretty much exactly what you said. I didn't realize portage is where the problem lies. In fact, I thought we were all still proud of portage I'm going to think about this some. - Grant -- [EMAIL PROTECTED] mailing list
[gentoo-user] Re: Gentoo Rules
Grant emailgrant at gmail.com writes: Multiple great ideas have already been suggested in this thread. Is this the first time they've been conceived and shared? Why hasn't work begun on them? Why isn't work completed on them? Because living costs money and Gentoo doesn't pay. I've been in business for 7 years and I'd like to take a shot at designing a system that would pay Gentoo developers to develop. Oh, I've taken this approach several times. My simple idea was to for an individual or company to solicit specific things they want, like a meta package for secure E-commerce. I'd 'Poney' up the funds and the greater community benefits. No takers. I'd throw a few thousand dollars at such a venture. I'm sure other would donate to. Anyone interested? (serious doubts among the dev ranks). Gentoo is a 'boys club' and any mention of using it formally to make money (for the greater gentoo community) sends the squirrels with their nuts running for cover. -- jaded after a few years, James -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
On Saturday 15 December 2007, Randy Barlow wrote: Yeah, that's the kinds of differences of opinion that are in the bug report, which is part of what makes this a more difficult ebuild to write. Things like libraries are really easy because it's just a configure make make install, but here you have a lot of configuration files and differences of opinion. I was thinking that a USE variable could be in order here, to support suid and a separate apache instance. Perhaps the variable could be suid? First, my opinion is that BackupPC, while being a wonderful application, should be split in two parts (by upstream): the daemon, and the web interface, so that they don't have to run on the same machine. Last time I checked, this was not supported, even if the config file seemed to allow for such a config. If this was the architecture, then it could be split into two distinct ebuilds, like eg zabbix or other apps. That said, and things being the way they are now, I'd make the suid behavior the default, since it requires less changes in a running system (a perl reemerge at most - assuming of course apache is already installed). If the user wants the separate apache instance, then he can set, say, USE=private-apache to get it. (btw, do gentoo initscripts support starting multiple instances of a daemon, perhaps under different users and using different parameters? I'd not bet on it, but I may be wrong. If it's not supported, waiting for baselayout to support this may take a long time, so it would be better to release the easier suid version in the meanwhile.) But again, there might be better arguments for doing the opposite. I'll take a look at the b.g.o. page where the ebuild is discussed (last time I checked was long ago, and things have surely evolved by now). -- [EMAIL PROTECTED] mailing list
[gentoo-user] Re: Gentoo Rules
b.n. brullonulla at gmail.com writes: I offered to take over the maintenance of the package and web installation page, and was turned down (probable by some punk under the age of 20) Sad. Can you link the thread? I think that would be counter productive. It's the 'culture of gentoo' that encourages this condescending attitude from the (document and other) devs. Lots of folks with skills would help, if the process of helping becomes lightweight or unencumbered. This sort of help could be outside of portage so anyone using such packages would be warned ahead of time that these packages are not officially sanctioned by the 'gentoo devs'. Maybe just a simple way to aggregate overlays and a place where others are encouraged to develop, use and maintain ebuilds and docs. Then the ordained among the gentoo devs could merely mine the repository for new additions to what they care to sanction. Right now, things are very discriminatory, discouraging others from participation and after awhile, folks leave or ignore the greater gentoo community. It's not the first time that there have been acrimony in the ranks of the gentoo devs.. imho, James -- [EMAIL PROTECTED] mailing list
[gentoo-user] Re: DMZ on an vmware gentoo guest running on winXP host
Mick [EMAIL PROTECTED] writes: On Friday 14 December 2007, [EMAIL PROTECTED] wrote: Mick [EMAIL PROTECTED] writes: Would I likely be opening my lan up for some christmas shopping by having a gentoo guest on a WinXP host running as a DMZ machine? It would be pretty barebones with a IPTABLE setup for logging and tagging or whatever I get interested in doing with the traffic. No X server or other frills. A rather simpler solution to do this would be to get hold of hub, connect it to the firewall and watch everything that passes through it. I do have an older hub, but not sure what you mean here. The hub has no network address and of course is not switched so anything going thru it can be filtered with tcpdump. But the router is switched. Not sure how a hub would see the outfacing address. I'd be able to see all the lan machines that were going thru it, but how about the traffic that the firewall is rejecting? Thats what I'm after. Can you elaborate a little? Maybe you mean something different by `hub'. I mean a hardware hub, not a switch and not a router. You need to place it in-line between your router/switch and your modem. Being on the WAN side of your NAT it will 'see' all the packets that go to/from the Internet (unfiltered). On the other side of the router you get the filtered traffic which when compared/contrasted with the WAN side will show you what the router and it's firewall are doing. I hope this is a bit clearer, otherwise please email me if you think this is getting off topic. I guess someone will squawk if they think it is not topical here, but it I think it should be ok since its about a specific setup involving a gentoo box or hardened VM gentoo guest. Below is a ascii art diagram of my simple network. I think you are talking about placing the hub as shown there. If I got that right then what I don't understand is how you talk to the hub. I mean if you connect it to any machine in the diagram or elsewhere wouldn't you be exposing that machine to the unfiltered internet? It still seems you would need somekind of hardened interface to that hub, but I'm probably not understanding how it would work.. ISP ISP ^ ^ | DSL Modem | X = hub | | -NetGearRouter/switch-- || | | || | | || | | -- -- - -- GentooWinXP WinXP WinXP -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: DMZ on an vmware gentoo guest running on winXP host
[EMAIL PROTECTED] wrote: I mean if you connect it to any machine in the diagram or elsewhere wouldn't you be exposing that machine to the unfiltered internet? I think that's the idea here - to see the difference between the two sides of the router. As far as getting that data safely in a virtual machine running on windows - I'd say having windows involved at all here is a risk. But you could definitely do it... -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: [Fwd: Re: Gentoo Rules]
Etaoin Shrdlu wrote: (btw, do gentoo initscripts support starting multiple instances of a daemon, perhaps under different users and using different parameters? I'd not bet on it, but I may be wrong. If it's not supported, waiting for baselayout to support this may take a long time, so it would be better to release the easier suid version in the meanwhile.) It's not too hard to start a separate instance of apache. You just copy /etc/init.d/apache2 to, say, /etc/init.d/backuppcApache2. Likewise copy the /etc/conf.d scripts, and change in the backuppc one the reference to the httpd.conf to, say, /etc/BackupPC/httpd.conf. Then, in that .conf file, make sure that you change the things to be suitable for BackupPC (in particular, get rid of the lines that include *.conf's from certain directories because these will cause apache to try and use the same PID! Make sure you specify a new PID file, among a few other related things) I really don't think the ebuild should let you use the same instance of apache that /etc/init.d/apache2 starts, because this would be a security risk. For example, I use BackupPC to back up three machines, in their entirety. That means that backuppc has the rights to change any files on those three machines. I've also got a webserver running, open to the internet, on my backuppc machine. If people on the internet can access backuppc, they can pretty much access all three of those other machines. But if I run on port 8080, and have that port blocked by a firewall, this is no longer a concern. The other option is to install password protection by default, but then you have to have competent users who can change the httpd passwords. I suppose you could write this as an instruction at the end of the ebuild. But, are htaccess passwords sent in plaintext? If so, that's also a major security risk. -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Kernel schedulers
On 15:27 Thu 13 Dec , Jason Carson wrote: Greetings, Where in the kernel config (make menuconfig) do I find the choice for schedulers. The one I am currently using is Anticipatory. What is the newest and latest scheduler for 2.6.23? Regards, Jason Carson -- [EMAIL PROTECTED] mailing list Like someone else mentioned, you can switch the sched on the fly, and quite easily. From what I have seen myself: Anticipatory seems to be, at times, faster than deadline, but not by much. It tries to predict what will be needed next, where as deadline makes reads/writes based on which will be the fastest (recomended for databases and such iirc). In my experiance, CFQ has always been the slowest. It gives everything even time, and seems to cause alot more head movement than the other two, which is a pain. Best bet is to compile them all in, and switch them out to see what works best. For me that seems to be deadline (btw, I am running a desktop), but testing would be the best thing. -- [EMAIL PROTECTED] mailing list
[gentoo-user] Re: DMZ on an vmware gentoo guest running on winXP host
Randy Barlow [EMAIL PROTECTED] writes: [EMAIL PROTECTED] wrote: I mean if you connect it to any machine in the diagram or elsewhere wouldn't you be exposing that machine to the unfiltered internet? I think that's the idea here - to see the difference between the two sides of the router. If that is the case then I guess I don't see how the quote below applies. From Mick in his initial reply: A rather simpler solution to do this would be to get hold of hub, connect it to the firewall and watch everything that passes through it. I relize you are not who made the reply I quote above but: If you still have to come up with a hardened interface to the hub then how is it simpler? Further, since the router is switched then you'd really need two hubs. One on each side, if the aim were to compare what is coming and what is getting thru. So we're getting further and futher away from `rather simpler' Come up with the hardened interface and forget the hub[s]. As I said my router offers to send all the bounced traffic to a designated DMZ. I am probably not interested enough right now to build up a whole different machine to talk to the hub or be the DMZ. So if you are pretty convinced doing it from a VMgentoo appliance running on one of the win boxes then I'll probably just keep fiddling around with the logs produced by the router. ... Thanks -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
On Sat, 15 Dec 2007 11:41:21 -0500, Randy Barlow wrote: Maybe his/her laptop doesn't stand the thermal output of its CPU when emerging or maybe he/she's the administrator of a large company's network, trying to move every computer system to Gentoo. Check out distccd! How does that help? Either every machine on the network spends time compiling when it should be earning its keep, or they all pass all the load to a compiler system that has to recompile the same code over and over for each computer on the network. Using that computer to build binary packages that the others can install is so much more efficient. -- Neil Bothwick Behaviorist psychology -- pulling habits out of rats signature.asc Description: PGP signature
[gentoo-user] emerging mythtv? permission denied...
Hi there!
I just tried to upgrade my mythtv installation, but the ebuild fails in the
unpack phase with a permission denied-error when accessing the
svn-repository:
Emerging (1 of 2) media-tv/mythtv-0.20.2_p14814 to /
* checking ebuild checksums ;-) ... [ ok ]
* checking auxfile checksums ;-) ... [ ok ]
* checking miscfile checksums ;-) ... [ ok ]
* This ebuild now uses a heavily stripped down version of your CFLAGS
* Don't complain because your -momfg-fast-speed CFLAG is being stripped
* Only additional CFLAG issues that will be addressed are for binary
* package building.
Unpacking source...
* subversion update start --
* repository:
http://svn.mythtv.org/svn/branches/release-0-20-fixes/mythtv
svn: Can't open file '.svn/lock': Permission denied
*
* ERROR: media-tv/mythtv-0.20.2_p14814 failed.
* Call stack:
* ebuild.sh, line 1701: Called dyn_unpack
* ebuild.sh, line 817: Called qa_call 'src_unpack'
* ebuild.sh, line 44: Called src_unpack
* mythtv-0.20.2_p14814.ebuild, line 108: Called subversion_src_unpack
* subversion.eclass, line 254: Called subversion_fetch
* subversion.eclass, line 189: Called die
* The specific snippet of code:
* ${ESVN_UPDATE_CMD} ${options} || die ${ESVN}: can't update
from ${repo_uri}.
* The die message:
* subversion.eclass: can't update from
http://svn.mythtv.org/svn/branches/release-0-20-fixes/mythtv.
*
* If you need support, post the topmost build error, and the call stack if
relevant.
* A complete build log is located
at '/var/tmp/portage/media-tv/mythtv-0.20.2_p14814/temp/build.log'.
*
zeus ~ # ls -l /var/tmp/portage/media-tv/mythtv-0.20.2_p14814/
total 0
drwxr-xr-x 2 portage root 48 Dec 15 20:38 distdir
drwxrwxr-x 3 portage portage 80 Dec 15 20:38 homedir
drwxrwxr-x 3 portage portage 168 Dec 15 20:38 temp
drwx-- 2 portage portage 48 Dec 15 20:38 work
zeus ~ #
Any suggestions?
Thanks in advance
Alex Puchmayr
--
[EMAIL PROTECTED] mailing list
[gentoo-user] ipw in kernel-genkernel-x86-2.6.22-gentoo-r9
Hi - Can anyone give me pointers on how to get ipw2100-firmware working with kernel-genkernel-x86-2.6.22-gentoo-r9? It works fine with kernel-genkernel-x86-2.6.22-gentoo-r8 where it's compiled in the kernel but the same set-up fails in kernel-genkernel-x86-2.6.22-gentoo-r9. When I run emerge ipw2100-firmware it seems to install OK, but the kernel can't seem to find the module. Thanks, Richard -- [EMAIL PROTECTED] mailing list
[gentoo-user] I don't understand what's happening with emerge gnupg
I just did an emerge --sync, and checked what was available for updateing... [m3000][root][~] emerge --ask --deep --update --world These are the packages that would be merged, in order: Calculating world dependencies... done! [ebuild U ] sys-devel/gnuconfig-20070724 [20070118] [ebuild U ] dev-libs/libassuan-1.0.2-r1 [0.6.10] [ebuild U ] sys-devel/gettext-0.17 [0.16.1-r1] [ebuild U ] dev-util/dialog-1.1.20071028 [1.1.20070930] [ebuild U ] sys-libs/com_err-1.40.3 [1.40.2] [ebuild U ] www-client/w3m-0.5.2-r1 [0.5.2] [ebuild U ] sys-libs/ss-1.40.3 [1.40.2] [ebuild U ] sys-fs/e2fsprogs-1.40.3 [1.40.2] [ebuild U ] net-firewall/iptables-1.3.8-r2 [1.3.8-r1] [ebuild U ] x11-apps/xinit-1.0.5-r1 [1.0.4] USE=-hal% -pam% [ebuild N] app-crypt/pinentry-0.7.3 USE=ncurses -caps -gtk -qt3 [ebuild U ] app-crypt/gnupg-2.0.7 [1.4.7-r1] USE=-doc% -openct% -pcsc-lite% [blocks B ] =app-crypt/gnupg-2.0.1 (is blocking app-crypt/gnupg-2.0.7) I do *NOT* have *ANY* packages masked in /etc/portage/package.mask Anyhow, I manually unmerged gnupg (would you believe version 1.4.7-r1 ?), and asked for an update again... [m3000][root][~] emerge --ask --deep --update --world These are the packages that would be fetched, in order: Calculating world dependencies... done! [ebuild U ] sys-devel/gnuconfig-20070724 [20070118] [ebuild U ] dev-libs/libassuan-1.0.2-r1 [0.6.10] [ebuild U ] sys-devel/gettext-0.17 [0.16.1-r1] [ebuild U ] dev-util/dialog-1.1.20071028 [1.1.20070930] [ebuild U ] sys-libs/com_err-1.40.3 [1.40.2] [ebuild U ] www-client/w3m-0.5.2-r1 [0.5.2] [ebuild U ] sys-libs/ss-1.40.3 [1.40.2] [ebuild U ] sys-fs/e2fsprogs-1.40.3 [1.40.2] [ebuild U ] net-firewall/iptables-1.3.8-r2 [1.3.8-r1] [ebuild U ] x11-apps/xinit-1.0.5-r1 [1.0.4] USE=-hal% -pam% [ebuild N] app-crypt/pinentry-0.7.3 USE=ncurses -caps -gtk -qt3 [ebuild N] app-crypt/gnupg-2.0.7 USE=bzip2 -doc -ldap -nls -openct -pcsc-lite (-selinux) -smartcard Curious minds want to know... - why didn't portage replace the old version itself? That's generally part of the update process. - once I manually unmerged gnupg, what brings it back? I did *NOT* use revdep-rebuild. - what's with the additional flags for 2.0.7 compared to what 2.0.7 had the previous time I tried? -- Walter Dnes [EMAIL PROTECTED] I'm not repeating myself I'm an X Window user... I'm an ex-Windows-user -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] I don't understand what's happening with emerge gnupg
On Sat, 15 Dec 2007 18:55:43 -0500, Walter Dnes wrote: - why didn't portage replace the old version itself? That's generally part of the update process. GnuPG is slotted, so 1.* and 2.* can be installed simultaneously. - once I manually unmerged gnupg, what brings it back? I did *NOT* use revdep-rebuild. Use --tree with the emerge command to see this. - what's with the additional flags for 2.0.7 compared to what 2.0.7 had the previous time I tried? Unless you use --verbose, emerge shows only the USE flags that have changed since the current install. As you no longer have GnuPG installed, that is now all flags. -- Neil Bothwick As a computer, I find your faith in technology amusing. signature.asc Description: PGP signature
Re: [gentoo-user] Gentoo Rules
On Fri, Dec 14, 2007 at 07:48:12AM -0800, Grant wrote Lately I've been shopping around for other distros as well as looking at *BSD. Gentoo development seems to have slowed way down and I like things being improved as quickly as possible. One item (amongst many) that chased me away from Windows was the upgrade treadmill... 1993 Win3.1 1994 Win3.11 1995 Win95 1996 Win95OSr2 1997 And on the 1997th year, Go^H^H Gates rested 1998 Win98 1999 Win98SE 2000 WinME and Win2K 2001 WinXP Remember that Gentoo is a linux distro, meaning that it is an implementation based on Linus Torvalds' baby. You may be thinking back to the early days of Kernel 2.0, 2.2, 2.4, etc, etc. For whatever reason, 2.6 has been around for 4 years, with no signs of being replaced. Gentoo can't do much about that, or do you want to fork the linux kernel? There have been a lot of minor improvements that, taken together, have led to major overall improvement. Basically, linux has caught up to Windows, indeed 64-bit linux has been better than 64-Windows for the past year. The concept is second to none, the execution of that concept is fantastic, but it needs to keep moving forward. Do you have a list of significant improvements you want implemented? There were gaping holes in linux's abilities in the past. These problems have been fixed. Change simply for the sake of change is a relic of the Windows era. -- Walter Dnes [EMAIL PROTECTED] I'm not repeating myself I'm an X Window user... I'm an ex-Windows-user -- [EMAIL PROTECTED] mailing list
[gentoo-user] dd questions
Hi group, I moved a partition(about 60G) from one drive to another, slightly larger, using: dd if=/dev/hdc4 of=/dev/hdc6 bs=32k When the operation completed this appeared: 904017 +1 records in 1904017 record out numbers which are precisely 999,999 apart. What is that all about? Followed by this: display all 902 possibilities (y or n) y and the screen filled up with an alphabetical list of executable programs like ls, something-something.sh etc. Which is strange because it was /home and contained no files like that. Google doesn't know, anybody here know? Just curious, doesn't seemt to be anything the matter with the new partition. Maxim Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] dd questions
[snip] Followed by this: display all 902 possibilities (y or n) y and the screen filled up with an alphabetical list of executable programs like ls, something-something.sh etc. Which is strange because it was /home and contained no files like that. This is what happens when you hit TAB twice when running bash. So either you hit TAB or a couple TAB characters \t\t were in the input stream somewhere. The list is a list of executable commands within the scope of your account. If you type part of a command and hit TAB twice, it'll give you a list of commands based on what you've typed so far. It's a little helper feature. Google doesn't know, anybody here know? Just curious, doesn't seemt to be anything the matter with the new partition. Maxim -Statux -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] realtek 8197 wireless card setup
On Tuesday 11 December 2007 09:48:10 am Mick wrote: On Tuesday 11 December 2007, Jeff Cranmer wrote: I believe that I have this enabled, however ieee80211 is still barfing out by asking for CONFIG_NET_RADIO. I'll check and confirm this tonight. Also check bugzilla. I remember reporting a bug with the more recent kernels failing to build kernel drivers. The last kernel that I managed to build rt2570 was 2.6.20-gentoo-r8. Kernel 2.6.23-gentoo-r3 fails to emerge any driver whatsoever. I have now (finally) successfully compiled the latest kernel 2.6.23-gentoo-r3 kernel. Once I enabled the 'Generic IEEE 802.11 Networking Stack (mac80211)' option in Networking -- Wireless, the Realtek RTL8187 USB support option appears in Device Drivers -- Network Device Support -- Wireless LAN section under wireless LAN (IEEE 802.11). With this RTL8187 driver compiled into the kernel, I get some success. Running the command 'dmesg | grep rtl8187' after reboot returns the message usbcore: registered new interface driver rtl8187 There is, however, no /etc/init.d/net.wlan0 on my system, so I'm not quite there yet. There is a net.eth0 (wired network), and a net.lo What do I need to do to get net.wlan0 active? Thanks Jeff -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] emerging mythtv? permission denied...
Alexander Puchmayr wrote: http://svn.mythtv.org/svn/branches/release-0-20-fixes/mythtv svn: Can't open file '.svn/lock': Permission denied Speaking out of ignorance here, but I wonder if .svn/lock is a file on your local machine, or if it's on the svn server? I know nothing about svn, other than it also does what cvs does :) -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Gentoo Rules
Neil Bothwick wrote: Maybe his/her laptop doesn't stand the thermal output of its CPU when emerging or maybe he/she's the administrator of a large company's network, trying to move every computer system to Gentoo. Check out distccd! How does that help? Either every machine on the network spends time compiling when it should be earning its keep, or they all pass all the load to a compiler system that has to recompile the same code over and over for each computer on the network. Using that computer to build binary packages that the others can install is so much more efficient. It does help in the case that Florian mentioned, also quoted at the top, about the laptop... I agree that if you are going to build the same thing for every computer it makes sense to do it just once. I was thinking more about the laptop... -- Randy Barlow http://electronsweatshop.com -- [EMAIL PROTECTED] mailing list
