Re: locate weirdness
guys, it was so funny to see you biting each other. come on, can you do it one more time, please ? 2012/1/23 Nico Kadel-Garcia > On Sun, Jan 22, 2012 at 5:38 PM, L. V. Lammert wrote: > > On Sun, 22 Jan 2012, Philip Guenther wrote: > > > > > >> > > There is no way of knowing if it would have found the problem, so why > > continue with this drivel? Contrary to the lengthy diatribes here trying > > to distract from the original problem an solution: > > > > 1) The problem with locate was traced to a bunch of session files; > > 2) The problem was fixed by cleaning them the hard way. > > > > There is no way to know if an upgrade would have fixed the problem, as > > upgrading is/was/would be just a distraction; it is not good practice to > > try and obscure the problem, and I do not understand why some people here > > like to expouse such practices. > > > > Sure, there is no support for 4.3, but, then I did not ASK for support on > > 4.3 (to read the OP). Don't bother to try and dixtract from the original > > problem - it juse makes it harder for those LOOKING for the problem and > > solution to find it in all the noise. > > As someone who's faced this kind of thing from both sides, I think > you're going to have a long term problem with the "just help me fix > the system I have, don't bother with telling me to upgrade" approach. > Too many bugs are fixed as part of re-engineering or feature addition, > and expecting even the authors, whom you are not paying for contracted > work, to maintain the old releases becomes futile pretty quickly. It's > difficult for them to maintain the old environments as test beds, or > to dredge back that far into memory of how things used to be done. > I've been running into this for decades, all the way back to the shift > from BSD 4.2 to BSD 4.3. (Note that that is not OpenBSD, it's BSD.) > > The yelling and namecalling is unfortunate. But from observation and > professional experience, if you want professional grade support for a > software livecycle of over 3 years, you should be willing to pay for > it.
Re: locate weirdness
2012/1/23 Lars : > Also MySQL became a billion dollar company and it doesn't even sell any > product http://www.mysql.com/products/ http://www.mysql.com/products/cluster/faq.html#20
Re: locate weirdness
On 23 January 2012 04:04, Lars wrote: > > It would be cool to see more BSD consulting companies and support > companies, and web hosting companies. If I was to look for an OpenBSD web > hosting company, I'd have difficulty finding one. There are far too many > LINUX web hosts out there. BSD seems like such a small tiny niche. The > market is saturated with LINUX web hosts and linux support companies, but > BSD is lacking. People often use Linux because it's the popular web > hosting O/S, not because it is a better tool. http://openbsd.org/support.html
Re: locate weirdness
Nico Kadel-Garcia wrote: ... > The yelling and namecalling is unfortunate. But from observation and > professional experience, if you want professional grade support for a > software livecycle of over 3 years, you should be willing to pay for > it. > > It would be cool to see more BSD consulting companies and support companies, and web hosting companies. If I was to look for an OpenBSD web hosting company, I'd have difficulty finding one. There are far too many LINUX web hosts out there. BSD seems like such a small tiny niche. The market is saturated with LINUX web hosts and linux support companies, but BSD is lacking. People often use Linux because it's the popular web hosting O/S, not because it is a better tool. Also MySQL became a billion dollar company and it doesn't even sell any product, it just offers "support" that you pay for. You wouldn't think support would gain you millions of dollars in sales, you would think mysql would only be financially successful if it sold the actual database product. Then again, we wouldn't want another fork of BSD where it was a separate distro that you had to pay for support for, like "Redhat BSD" or something cheesy like that.
Re: locate weirdness
On Sun, Jan 22, 2012 at 5:38 PM, L. V. Lammert wrote: > On Sun, 22 Jan 2012, Philip Guenther wrote: > > >> > There is no way of knowing if it would have found the problem, so why > continue with this drivel? Contrary to the lengthy diatribes here trying > to distract from the original problem an solution: > > 1) The problem with locate was traced to a bunch of session files; > 2) The problem was fixed by cleaning them the hard way. > > There is no way to know if an upgrade would have fixed the problem, as > upgrading is/was/would be just a distraction; it is not good practice to > try and obscure the problem, and I do not understand why some people here > like to expouse such practices. > > Sure, there is no support for 4.3, but, then I did not ASK for support on > 4.3 (to read the OP). Don't bother to try and dixtract from the original > problem - it juse makes it harder for those LOOKING for the problem and > solution to find it in all the noise. As someone who's faced this kind of thing from both sides, I think you're going to have a long term problem with the "just help me fix the system I have, don't bother with telling me to upgrade" approach. Too many bugs are fixed as part of re-engineering or feature addition, and expecting even the authors, whom you are not paying for contracted work, to maintain the old releases becomes futile pretty quickly. It's difficult for them to maintain the old environments as test beds, or to dredge back that far into memory of how things used to be done. I've been running into this for decades, all the way back to the shift from BSD 4.2 to BSD 4.3. (Note that that is not OpenBSD, it's BSD.) The yelling and namecalling is unfortunate. But from observation and professional experience, if you want professional grade support for a software livecycle of over 3 years, you should be willing to pay for it.
Re: locate weirdness
Lammert, I am going to have to ask you to leave our lists. You are no longer welcome here.
Re: locate weirdness
On Sun, 22 Jan 2012, Philip Guenther wrote: > There is no way of knowing if it would have found the problem, so why continue with this drivel? Contrary to the lengthy diatribes here trying to distract from the original problem an solution: 1) The problem with locate was traced to a bunch of session files; 2) The problem was fixed by cleaning them the hard way. There is no way to know if an upgrade would have fixed the problem, as upgrading is/was/would be just a distraction; it is not good practice to try and obscure the problem, and I do not understand why some people here like to expouse such practices. Sure, there is no support for 4.3, but, then I did not ASK for support on 4.3 (to read the OP). Don't bother to try and dixtract from the original problem - it juse makes it harder for those LOOKING for the problem and solution to find it in all the noise. Lee
Re: locate weirdness
On Sun, Jan 22, 2012 at 12:39 PM, L. V. Lammert wrote: > On Sun, 22 Jan 2012, Amit Kulkarni wrote: >> a recent system does a fsck -fp of each partition, so it would have >> fixed your problem. >> > Would 4.4? 4.5? I don't think so, You keep stating things that you believe to be true without doing any checking of them. This one is false. > .. if you recall, the system is 4.3 so a > normal upgrade (i.e. not a clean install) would not have fixed the > problem. You apparently have never done an upgrade or don't remember what it did when you last did so. Amit is correct that doing an upgrade from 4.3 or 4.4 or 4.5 would have, by default, done a forced fsck of your filesystems. If you're interested in bringing your beliefs closer to reality, then I suggest you go pick up a copy of 4.4 or 4.5 and try the upgrade, or read the source for the upgrade script. Philip Guenther
Re: locate weirdness
On Sun, 22 Jan 2012, Amit Kulkarni wrote: > a recent system does a fsck -fp of each partition, so it would have > fixed your problem. > Would 4.4? 4.5? I don't think so, .. if you recall, the system is 4.3 so a normal upgrade (i.e. not a clean install) would not have fixed the problem. > you are very ungrateful, by insulting a person who helped you, for free. > Sorry, Insults beget Insults - that's the OBSD way, is it not? If you would take a little less effort to divert the original question, we would all have a much nicer experience. Lee
Re: locate weirdness
>> For those that happen to google this thread trying to find the >> solution: upgrading *would* have fixed his system due to the various >> steps done during upgrades, >> > BZZZT! WRONG! If the system rebooted clean [which this one did], the > problem would not have been found during a normal upgrade. > > Again, misdirection, tons of rheotiric deflecting the issues, and normal > behavior of many on this list DID totally obscure the problem. *BUT* > that's the way it works here. a recent system does a fsck -fp of each partition, so it would have fixed your problem. you are very ungrateful, by insulting a person who helped you, for free.
Re: locate weirdness
On Sun, 22 Jan 2012, Philip Guenther wrote: > And yet you, L.V. Lammert, use the code, don't clearly explain what > was failing on your system, and behave in exactly the way you rant > about. > Now THAT's quite laughahble! The ONLY problem here is with all the IDIOTS spouting CRAP with no interest whatsoever in solving the PROBLEM that was expressed. > For those that happen to google this thread trying to find the > solution: upgrading *would* have fixed his system due to the various > steps done during upgrades, > BZZZT! WRONG! If the system rebooted clean [which this one did], the problem would not have been found during a normal upgrade. Again, misdirection, tons of rheotiric deflecting the issues, and normal behavior of many on this list DID totally obscure the problem. *BUT* that's the way it works here. Lee
Re: locate weirdness
On Sun, 22 Jan 2012, Anonymous wrote: > Whooever told him to man up was justified. > HUH?? WTF?? That's a CROCK!! It's amuzing to watch the flames flying, however, there *ARE* some people on the list that have an interest in solving problems. It IS unfortunate, however, that the actual PROBLEM and RESOLUTION may be hard to identify through all the noise. Lee
Re: locate weirdness
> And yet you, L.V. Lammert, I am not Lammert. I don't approve of the way Lammert handled the issue but that is separate from the fairly regular piss parades and walls of text taking place here when a simple answer would work. The consensus of the vocal parties is that contribution = abuseLicense and I call bullshit. Lammert SHOULD have apoligized to the list for fucking himself and wasting people's time but he was less likely to do that after Henning and others gratuitous ass parade and ensuing gang bang. Whoever told him to man up was justified.
Re: locate weirdness
On Sun, Jan 22, 2012 at 3:35 AM, Anonymous wrote: > I asked this before but I guess you didn't see it. So if you contribute > "much more code to OpenBSD" than someone else do you automatically get > license to insult people and post 100% noise as some kind of reward? Since you're such an incredibly brave man and used an anonymous email I don't know who the hell you are. Fritz? > Lars, you ass-licking dog, what I am saying to you and prima donnas like you > is you can be a good human being and that is more important than all the > patches and code in the world. If you can contribute patches and new code so > much the better, but if all you do is contribute to OpenBSD and you behave > like a fucking asshole you wipe out all the benefit. Cause the world does > need good human beings but it doesn't need prima donnas just because they > contribute to OpenBSD. I'm pretty sure the project would still be doing fine > even if acclaimed contributors and their ass-licking dog fanboys like > you didn't spend entire threads bashing people when a simple answer would be > enough. They know everything already right? so it should be easy to answer. > Fuck you and your boyfriend. All that talk about what matters and then you try to insult me by insinuating that I am gay. The true hallmark of a good human being, right? Good work on making the world a better place. Cheers, Lars
Re: locate weirdness
On Sat, Jan 21, 2012 at 11:35 AM, Anonymous wrote: > Lars, you ass-licking dog, what I am saying to you and prima donnas like you > is you can be a good human being and that is more important than all the > patches and code in the world. <...> And yet you, L.V. Lammert, use the code, don't clearly explain what was failing on your system, and behave in exactly the way you rant about. You actions demonstrate that, actually, you find the code more important than the behavior of the people that wrote it. For those that happen to google this thread trying to find the solution: upgrading *would* have fixed his system due to the various steps done during upgrades, such as fsck being run and permissions being corrected when the sets are untarred. Philip Guenther
Re: locate weirdness
Lars Hansson whined > > I notice you spend much more time scolding people than actually saying> > anything worthwhile. You should work on yourself and find out why that> > is. Perhaps you could benefit from some anger management training? > I notice that Henning is contributing much more code to OpenBSD than > you ever have and has also produces much more informative and useful > replies than you ever have. You should stop trolling and get a life. I asked this before but I guess you didn't see it. So if you contribute "much more code to OpenBSD" than someone else do you automatically get license to insult people and post 100% noise as some kind of reward? Lars, you ass-licking dog, what I am saying to you and prima donnas like you is you can be a good human being and that is more important than all the patches and code in the world. If you can contribute patches and new code so much the better, but if all you do is contribute to OpenBSD and you behave like a fucking asshole you wipe out all the benefit. Cause the world does need good human beings but it doesn't need prima donnas just because they contribute to OpenBSD. I'm pretty sure the project would still be doing fine even if acclaimed contributors and their ass-licking dog fanboys like you didn't spend entire threads bashing people when a simple answer would be enough. They know everything already right? so it should be easy to answer. > Cheers, > Lars Fuck you and your boyfriend.
Re: locate weirdness
> I notice you spend much more time scolding people than actually saying> > anything worthwhile. You should work on yourself and find out why that> is. > Perhaps you could benefit from some anger management training? I notice that Henning is contributing much more code to OpenBSD than you ever have and has also produces much more informative and useful replies than you ever have. You should stop trolling and get a life. Cheers, Lars
Re: locate weirdness
L. V. Lammert wrote on Thu, Jan 19, 2012 at 09:20:44AM -0600: > On Thu, 19 Jan 2012, Paul de Weerd wrote: >> With apologies to all, this will be my last reply on this thread. > Really? That WOULD be nice. Hopefully you will abide by your promise. Sure, I'd appreciate if Paul wrote a bit more, in particular in his capacity as an undeadly editor, where he contributed a few nice articles in the past. Always welcome on misc@, too, of course, especially in other threads... Yours, Ingo P.S. .procmailrc edited, finally.
Re: locate weirdness
On Thu, 19 Jan 2012, Paul de Weerd wrote: > With apologies to all, this will be my last reply on this thread. > Really? That WOULD be nice. Hopefully you will abide by your promise. Lee
Re: locate weirdness
On Wed, Jan 18, 2012 at 10:15 AM, L. V. Lammert wrote: > It was truly a shame that so many people here prefer to start their > flamethrowers rather than offer any sort of constructive information! In > this case, THE PROBLEM HAS BEEN IDENTIFIED AND IT WOULD MOST LIKELY NOT HAVE > BEEN FIXED WITH A NORMAL UPGRADE! !! !!! > > Before reading further, please REREAD the statement above. > > As it turns out, there WERE some folks here that had excellent suggestions > [privately], and that helped significantly in isolating the problem. So, posting to the list worked ;-) > Thanks > to who got me back on topic and reminded me of this possible issue, .. > and who suggested the simplest solution: > > find / ! \( -fstype ffs -or -fstype ufs -or -fstype ext2fs \) -prune -or > -path /tmp -prune -or -path /var/tmp -prune -or -path /usr/tmp -prune -or > -print > /tmp/locate test > > Running the find separately identified the file system problem, and it was > easily fixed as a result. > >Lee
Re: locate weirdness
With apologies to all, this will be my last reply on this thread. On Wed, Jan 18, 2012 at 04:18:11PM -0600, L. V. Lammert wrote: | At 03:59 PM 1/18/2012, you wrote: | >Wait - so there's an issue that you have identified, with help from | >members on this list ? And you're refusing to divulge the exact | >details that would probably help resolve the problem in future OpenBSD | >releases ? | | Of course, an exposition was to be expected, ... an off topic one | that proves that you don't read what I posted! Really ? You're going to state that I don't read what you posted ? Are you absolutely sure ? In the reply I sent on the 15th where I quote directly from your e-mail and spend 3 paragraphs on how you've confused the people you want to help you, where I *again* try to extract an answer from you .. this is where I did not read what you posted ? Note that you never bothered to reply to any of my questions. Take take take, because you're so entitled to free help. No giving; why should you ? Maybe you're talking about how you were able to find the problem with the find command you posted. That does not give any details about the problem (i.e. what you did wrong to cause the effects you were asking about), just how you found it in your case. Or are you perhaps referring to the fact that you failed to include the details I ask for in your first update to the list after you fixed the situation ? But that you did answer Peter Hessler's query for more details ? Because, yes, I also read that e-mail. But that was in reply to a guy asking for details - you know, what I was doing. Why didn't he get the same verbal ass whooping I got ? Because he beat me to it ? Or because my MX had not received your answer to him yet when I started my reply ? Or are you referring to the 'detailedness' of your reply ? You know, I was asking for "exact details". And you answer with (I quote again, more proof I don't read your e-mails, I suppose): > No, a "file system problem". In this case, cross-linked files from a > Rails application. We had already gathered that you had a (self inflicted) "file system problem". That's not really new information (although it's good that you finally realized it's actually *your* fault, not some bug in OpenBSD that was at fault here). So we now get to deal with your cryptic "cross-linked files from a Rails application". What the hell is that supposed to mean ? What are cross-linked files ? So there was a filesystem issue that you were able to fix with fsck(8) ? How does the Rails application come into play ? Only machines running Rails apps can suffer from this ? At least it corrects your answer to the "what did you change" question from "nothing" to "yeah, I was cross-linking files". And you think my reply was 'off topic'. So trying to find out if there's a bug in OpenBSD and wanting it fixed is off topic ? Trying to get you to do what you were advocating yourself in reply to Philip with your "Amen! At least there's a chance it would turn up in the search engines."; you know, trying to get you to be more detailed and more on-topic is off topic ? | If you check the rest of the thread, you will see that I did post | the exact cause; more details I will not provide as what's there is | sufficient to describe the problem and any *more* detail would just | be flame fodder. Yeah, I full well realize that you managed to fuck up your filesystem all by yourself causing the issues you've been giving other people grief over. I think everybody realizes that. Quite a few realized this from day one. Man up and detail what you did to get to this situation. That's not going to be flame fodder - "yeah people, sorry, my bad .. here's what I did wrong". We see those on the list quite often, and I don't think they are the big flame fodder you claim them to be. You being a prick about it *is* flame fodder. | >That's a great thank you to all those people that helped you, Lee, | >especially the ones you don't mention by name here below. Just great. | | Well, if you DO want credit, thanks for the wisdom hidden after the | exegis! (I do not name names without permission, *especially* on | this list.) Sorry, I'm not a native English speaker, so I have no idea what 'exegis' is, nor does my dictionary :( And yes, I was trying to help you. Mostly trying to help you help yourself. The one thing that is comforting in all this, is that I expect others on this list have more clue in helping themselves; whenever they run into a similar issue, they'll actually be able to find the problem and fix it all by themselves. And even if they don't, they'll probably be less of a prick then you've been through all this and simply explain what they fucked up on the list, gaining them some credits for fessing up to what is probably minor stupidity or just simply misunderstanding of the workings of their tools ("locate.updatedb does not run as root; wow, I didn't know that"). I'll be sure never to help
Re: locate weirdness
At 03:59 PM 1/18/2012, you wrote: Wait - so there's an issue that you have identified, with help from members on this list ? And you're refusing to divulge the exact details that would probably help resolve the problem in future OpenBSD releases ? Of course, an exposition was to be expected, ... an off topic one that proves that you don't read what I posted! If you check the rest of the thread, you will see that I did post the exact cause; more details I will not provide as what's there is sufficient to describe the problem and any *more* detail would just be flame fodder. That's a great thank you to all those people that helped you, Lee, especially the ones you don't mention by name here below. Just great. Well, if you DO want credit, thanks for the wisdom hidden after the exegis! (I do not name names without permission, *especially* on this list.) Good that you have at least taken the time and effort to make us reread your own little flame (in all caps, because that helps so much to bring the message across). Confirmed - thanks! Lee
Re: locate weirdness
Wait - so there's an issue that you have identified, with help from members on this list ? And you're refusing to divulge the exact details that would probably help resolve the problem in future OpenBSD releases ? What the hell. That's a great thank you to all those people that helped you, Lee, especially the ones you don't mention by name here below. Just great. Good that you have at least taken the time and effort to make us reread your own little flame (in all caps, because that helps so much to bring the message across). Paul 'WEiRD' de Weerd PS: Best part ? "MOST LIKELY" On Wed, Jan 18, 2012 at 10:15:34AM -0600, L. V. Lammert wrote: | It was truly a shame that so many people here prefer to start their | flamethrowers rather than offer any sort of constructive | information! In this case, THE PROBLEM HAS BEEN IDENTIFIED AND IT | WOULD MOST LIKELY NOT HAVE BEEN FIXED WITH A NORMAL UPGRADE! !! !!! | | | Before reading further, please REREAD the statement above. | | As it turns out, there WERE some folks here that had excellent | suggestions [privately], and that helped significantly in isolating | the problem. Thanks to who got me back on topic and reminded me | of this possible issue, .. and who suggested the simplest | solution: | | find / ! \( -fstype ffs -or -fstype ufs -or -fstype ext2fs \) -prune | -or -path /tmp -prune -or -path /var/tmp -prune -or -path /usr/tmp | -prune -or -print > /tmp/locate test | | Running the find separately identified the file system problem, and | it was easily fixed as a result. | | Lee | -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: locate weirdness
On Wed, 18 Jan 2012, Peter Hessler wrote: > On 2012 Jan 18 (Wed) at 10:15:34 -0600 (-0600), L. V. Lammert wrote: > :Running the find separately identified the file system problem, and > :it was easily fixed as a result. > > So, what was the actual problem? Permissions? > No, a "file system problem". In this case, cross-linked files from a Rails application. Lee
Re: locate weirdness
On Wed, 18 Jan 2012, Theo de Raadt wrote: > > On 2012 Jan 18 (Wed) at 10:15:34 -0600 (-0600), L. V. Lammert wrote: > > :Running the find separately identified the file system problem, and > > :it was easily fixed as a result. > > > > So, what was the actual problem? Permissions? > > Idiot near the keyboard. > Guess it takes one to try and identify another! Seriously, I have nothing but respect for the Dev team, .. which is why the OP was "has anyone ever seen". I did not EXPECT support from Dev's, so, to some degree, your insistance on upgrading was totally OT and not worth posting. Someone may experience a similar problem in the future, .. and as was mentioned, others have in the past. The solution posted may be of help in such cases. Lee
Re: locate weirdness
> On 2012 Jan 18 (Wed) at 10:15:34 -0600 (-0600), L. V. Lammert wrote: > :Running the find separately identified the file system problem, and > :it was easily fixed as a result. > > So, what was the actual problem? Permissions? Idiot near the keyboard.
Re: locate weirdness
On 2012 Jan 18 (Wed) at 10:15:34 -0600 (-0600), L. V. Lammert wrote: :Running the find separately identified the file system problem, and :it was easily fixed as a result. So, what was the actual problem? Permissions? -- There are two types of people in this world, good and bad. The good sleep better, but the bad seem to enjoy the waking hours much more. -- Woody Allen
Re: locate weirdness
It was truly a shame that so many people here prefer to start their flamethrowers rather than offer any sort of constructive information! In this case, THE PROBLEM HAS BEEN IDENTIFIED AND IT WOULD MOST LIKELY NOT HAVE BEEN FIXED WITH A NORMAL UPGRADE! !! !!! Before reading further, please REREAD the statement above. As it turns out, there WERE some folks here that had excellent suggestions [privately], and that helped significantly in isolating the problem. Thanks to who got me back on topic and reminded me of this possible issue, .. and who suggested the simplest solution: find / ! \( -fstype ffs -or -fstype ufs -or -fstype ext2fs \) -prune -or -path /tmp -prune -or -path /var/tmp -prune -or -path /usr/tmp -prune -or -print > /tmp/locate test Running the find separately identified the file system problem, and it was easily fixed as a result. Lee
Re: locate weirdness
On Thu, 12 Jan 2012 02:15:39 -0700 Duncan Patton a Campbell wrote: > On Wed, 11 Jan 2012 13:47:48 -0600 > "L. V. Lammert" wrote: > > > At 01:30 PM 1/11/2012, Jeremy O'Brien wrote: > > > > >4.3 was released May 1, 2008. That's almost 4 years old software. What > > >are you expecting here? Someone to check out the code from that > > >version and deeply inspect what may be causing your problem, that is > > >more than likely already fixed in a later version? > > > > Another typical reply - the question was "has anyone ever seen > > anything like this", .. or, perhaps, "what could be causing it". No > > need for the off-topic diatribes - a simple no would more than suffice. > > > > Lee > > > Yes. I have seen problems with locate in all releases that I have run. > Unfortunately I have never bothered to characterize these problems. > > Dhu > I should add that this has happened on deeply nested file systems with longpathnames and highbit characters. Dhu
Re: locate weirdness
> Perhaps he did. Wouldn't it be useful to help the guy trying to help > you (you know, the "wheat") by giving a really simple and > straightforward answer, even if it is repeating yourself ? Probably > would've been less typing than what you just did (e.g. "Sorry, I think > it runs as user "). No, we've been over that and the answer from the argumentative sonsabitches was "as long as you contribute one patch you are permitted nay might we suggest *encouraged* to flame rather than help." I guess it makes them feel like men, but to the rest of us they only seem like girl scouts. If you point this out you spawn an entire new subthread of postings by the abovementioned argumentative sonsabitches reaffirming their girl scout status. Why actually answer a question when you can create a sharkfest of insults and make yourself feel like a man? So what if you "contribute"? I fired a few assholes like you. I don't need prima donnas, you're not worth it. Helpful people who actually know something aren't mutually exclusive. To you primma donnas, go fuck yourselves. You aren't worth it.
Re: locate weirdness
On Sat, Jan 14, 2012 at 10:07:15PM -0600, L. V. Lammert wrote: | On Sat, 14 Jan 2012, Philip Guenther wrote: | | > Any progress? I see plenty of replies to the people that you *don't* | > think are helping you but no reply to my question about what user you | > think locate.updatedb runs as, something which does factor into being | > able to solve this... | > | The answer was already posted, .. perhaps you missed it? Perhaps he did. Wouldn't it be useful to help the guy trying to help you (you know, the "wheat") by giving a really simple and straightforward answer, even if it is repeating yourself ? Probably would've been less typing than what you just did (e.g. "Sorry, I think it runs as user "). However, unless I've missed an e-mail from you, you have *not* answered the question. Here's what you did reply (that is somewhat related to Philip's question): In > Agreed, .. but if locate.update does NOT run as root, that would seem to > indicate some problem other than permissions. ...and... In > If it does not run as root, then it isn't a permission issue as running as > root provides all required permissions, eh? > > I have never seen locate.updatedb fail when run as root (3.0 to 5.0, > actually), .. but, then, it isn't exactly 'failing', it just isn't > indexing anything except "/home". You see how you don't mention what user you think it runs as ? You make some roundabout statements concerning what would or would not work if it did or did not run as root (quite confusing), but a clear cut, straightforward answer is missing. Let's analyze one part in particular: "If it does not run as root, then it isn't a permission issue" If you do not run as root, you don't have root privileges so you have fewer permissions. But if you have these fewer permissions, that can not be an issue. Do you see how that statement is wrong ? Philip called you out on it, saying he didn't understand what you were saying. He replies with: In > I'm sorry, but I don't understand that sentence. It appears to > conflate running as root with not running as root, or I'm miscounting > the 'not's. > > So let me try again: what user do you think locate.updatedb is run as? Yet you never (publically) answered this e-mail (at least not according to http://marc.info/ or my own archive of misc@). Here's another piece of the puzzle for you .. locate.updatedb does NOT run as root by default: --- from /etc/weekly - UPDATEDB="/usr/libexec/locate.updatedb" echo "${UPDATEDB} --fcodes=- --tmpdir=${TMPDIR:-/var/tmp}" | \ nice -5 su -m nobody 2>/dev/null 1>$TMP -- So /by default/ it runs as user 'nobody'. Now you can change your setup so that locate.updatedb does run as root, but we all know you didn't do this as you would have mentioned it in your first post. One thing you actually could try is removing the 2>/dev/null from those lines, to see if any errors show up that might further help you debug this issue. Anyway, I hadn't seen a reply to my questions either. I'm quite curious by now what the find and locate snippets I suggested would output on your system. Also wondering if fsck showed you any errors which might explain what's going on. So, what's up with those ? Curious minds want to know. The archive is also still waiting for answers to provide as feedback to future google queries. Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: locate weirdness
On Sat, 14 Jan 2012, Philip Guenther wrote: > Any progress? I see plenty of replies to the people that you *don't* > think are helping you but no reply to my question about what user you > think locate.updatedb runs as, something which does factor into being > able to solve this... > The answer was already posted, .. perhaps you missed it? Lee
Re: locate weirdness
On Fri, Jan 13, 2012 at 1:02 PM, L. V. Lammert wrote: > Thanks again to those that actually read the original question, .. I am > continuing to try and resolve the issue. Any progress? I see plenty of replies to the people that you *don't* think are helping you but no reply to my question about what user you think locate.updatedb runs as, something which does factor into being able to solve this... Philip Guenther
Re: locate weirdness
On 2012-01-13, L. V. Lammert wrote: > *Especially* in this case since locate is a standard utility with a shell > script that has not changed between 4.3 & 4.9, and I expect it hasn't for > 5.0 either. If the system utilities have not changed, then the problem > must be elsewhere; Oh, c'mon, there have been 750-odd commits to libc in that timeframe and we have replaced the C compiler. If instead of writing snarky mails you had spent the time upgrading, by now we would know if it still affects -current and thus worth spending time investigating further.
Re: locate weirdness
On Fri, Jan 13, 2012 at 03:02:35PM -0600, L. V. Lammert wrote: | > Personally and If I had time I'd want to find out the problem but I'd be | > wiping and reinstalling from scratch anyway, especially with an unknown | > cause. | > | So, which is it? The attitude 'an upgrade will fix everything' is really | pretty dumb [though the core folks are certainly justfied as the problem | is most like not an issue for ongoing development, but that wasn't the | original question, was it?]. An upgrade doesn't fix everything, nobody said it would. What would be interesting is verifying whether an upgrade fixes the problem. It would bring the topic in the realm of supported releases and from there it may get more attention. | If something isn't working properly, throwing it away is ***NOT*** the | best solution! Would you take your car to the junkyard just because you | have a dome light that isn't working? You wouldn't download a car. This analogy with cars didn't work for the entertainment industry, and it doesn't work for you. Upgrading (especially in the case of OpenBSD) does not incur a cost outside of the time you need to spend on the upgrade itself (time you're now spending on debugging and e-mailing the list). You say you can't upgrade for reasons you don't want to divulge, so be it. But comparing this to taking a car to the junkyard makes no sense. And your nonsensical argument doesn't help your case. Here's another reason why your argument fails, btw. Replace your dome light; done. You know what the issue is and you can easily fix it (or have someone fix it for you). Quite different from your problem with locate / updatedb where you have no idea why it doesn't work. | *Especially* in this case since locate is a standard utility with a shell | script that has not changed between 4.3 & 4.9, and I expect it hasn't for | 5.0 either. If the system utilities have not changed, then the problem | must be elsewhere; blowing away a system just because you can't find the | problem is just plain stupid. I thought you already know what the problem is: locate doesn't work for you. What you want is to do is resolve this problem. If reinstalling the system resolves the issue, then why are you arguing ? Oh, you want to understand the problem ? Fair enough. So what you needed (and got) was help on how to debug problems. Sorry, Lee, but you've been on this list for a long time (you mentioned you were expecting some of the feedback you were less happy with from your experience on the list); personally I had expected you would be able to help yourself in this area. At the very least, it would've been prudent if you had provided details in your first e-mail that had to be extruded from you over the course of what is now quite a long thread (with many off-topic replies from yourself, by the way; shouldn't you be debugging an issue instead of arguing on the internet ?). | Thanks again to those that actually read the original question, .. I am | continuing to try and resolve the issue. You're honing skills, that's excellent. Just a bit of a shame you have so much issues with the "less useful" feedback you're getting. It's funny that you would answer a question about "what did you change" with 'nothing' when it relates to a filesystem. Are you absolutely, 100% positive you did not create any new files ? Or delete some ? If I were you, I'd try to look for files with weird names in your filesystems. Could they have been created recently, perhaps ? Also, try to figure out if the database really ends at the border between two partitions (/home and whatever comes next). Is *everything* from /home in the database ? find /home | while read FILENAME do RES=`locate "${FILENAME}" | head -n 1` [ "X${RES}" = "X" ] && echo Could not find ${FILENAME} done Realize that this provides false positives for files created since the last updatedb run; make sure to filter those out. Also, any errors displayed while you run this could be of interest... Also try to establish if it's really true that only files from /home are in the database: locate / | grep -v ^/home Finally - are your filesystems OK ? Force an fsck run. Not a guarantee that if you have issues in that area that they are found, but if it finds something that would also be telling. Again, sorry Lee, but this is basic stuff I would've expected you to provide in your first post. Help people help you, even if you're running code that's several years old. *Especially* if you're running code that's several years old - you know the mantra here is "upgrade to a supported release first", so try to put some effort in yourself (and show that you have). Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: locate weirdness
On Thu, 12 Jan 2012, Kevin Chadwick wrote: > I'm surprised you've had so much help. > You shouldn't be, .. there *ARE* a few decent folks here on the list. > Personally and If I had time I'd want to find out the problem but I'd be > wiping and reinstalling from scratch anyway, especially with an unknown > cause. > So, which is it? The attitude 'an upgrade will fix everything' is really pretty dumb [though the core folks are certainly justfied as the problem is most like not an issue for ongoing development, but that wasn't the original question, was it?]. If something isn't working properly, throwing it away is ***NOT*** the best solution! Would you take your car to the junkyard just because you have a dome light that isn't working? *Especially* in this case since locate is a standard utility with a shell script that has not changed between 4.3 & 4.9, and I expect it hasn't for 5.0 either. If the system utilities have not changed, then the problem must be elsewhere; blowing away a system just because you can't find the problem is just plain stupid. Thanks again to those that actually read the original question, .. I am continuing to try and resolve the issue. Lee
Re: locate weirdness
On Wed, 11 Jan 2012 20:28:57 +0100 Marian Hettwer wrote: > Try to look from a different angle here. > Say, you would have an old Debian Sarge release (years old) and you > would approach a debian mailing list with "something is weird with > locate", pretty sure you would get a lot of advises to upgrade first, > test then, and if problem persists, come back. >From the blogs I've read where this happens it's far harsher than that, due to the bugs in Linux. Something like "what the fsck are you doing running a system that old your a danger to the internet. It's people like you that.." I've never seen a come back like "why's this Linux kernel such a danger to the world in just 2 months then?, every router on this street is running a > year old firmware! and the current firmware released 4 weeks ago has 2 critical bugs described as 'just another bug'" I'd love to see the replies if any to that. But this is OpenBSD and I assume your just running a simple firewall or on a closed network or know your systems services are "known bug" free. I'm surprised you've had so much help. Personally and If I had time I'd want to find out the problem but I'd be wiping and reinstalling from scratch anyway, especially with an unknown cause. Of course having install scripts makes that decision much easier. It shouldn't be hard to copy your configs off, just make a root drive backup first in case you miss something. Surely faster than reading the upgrade guides for 7 releases.
Re: locate weirdness
On 01/13/12 13:50, Fritz Wuehler wrote: Spoke the self-proclaimed guru: Upgrading is a rule of this list. It cannot get anymore simple than that. So you're saying OpenBSD and Windows are really the same? No need to actually diagnose problems just upgrade. Whatever it is is fixed in the current version. Arsewipe LOL Yes. For old stuff, that's exactly what we say. Why should we chase bugs that possibly have already been fixed, either specifically or as a consequence of four years of development? Upgrade, and if it's still there, someone might be interested in helping out. So now just accept it or help the guy out yourself. Off-list. /Alexander
Re: locate weirdness
On 01/13/12 14:47, Otto Moerbeek wrote: On Fri, Jan 13, 2012 at 08:38:37AM -0500, Kenneth R Westerback wrote: On Thu, Jan 12, 2012 at 03:28:37PM +0100, Fritz Wuehler wrote: Hey Henning, off-topic diatribes? coming to this mailing list asking for help about a 4 year old release when it is clearly documented that you are ENTIRELY ON YOUR OWN with releases older than a year is at least off topic, if not outright rude. I notice you spend much more time scolding people than actually saying anything worthwhile. You should work on yourself and find out why that is. Perhaps you could benefit from some anger management training? so my advice is to upgrade. You could have said just that without the assholier than thou 'tude. I know, you can't help it. Something you and he apparently have in common. henning balances it out by contributing code. What have you contributed? Ken He balances out by giving free personal consultation. -Otto insultation you say?
Re: locate weirdness
Spoke the self-proclaimed guru: > Upgrading is a rule of this list. It cannot get anymore simple than that. So you're saying OpenBSD and Windows are really the same? No need to actually diagnose problems just upgrade. Whatever it is is fixed in the current version. Arsewipe LOL
Re: locate weirdness
On Fri, Jan 13, 2012 at 08:38:37AM -0500, Kenneth R Westerback wrote: > On Thu, Jan 12, 2012 at 03:28:37PM +0100, Fritz Wuehler wrote: > > Hey Henning, > > > > > off-topic diatribes? coming to this mailing list asking for help about > > > a 4 year old release when it is clearly documented that you are > > > ENTIRELY ON YOUR OWN with releases older than a year is at least off > > > topic, if not outright rude. > > > > I notice you spend much more time scolding people than actually saying > > anything worthwhile. You should work on yourself and find out why that > > is. Perhaps you could benefit from some anger management training? > > > > > so my advice is to upgrade. > > > > You could have said just that without the assholier than thou 'tude. I know, > > you can't help it. > > > > Something you and he apparently have in common. > > henning balances it out by contributing code. What have you > contributed? > > Ken He balances out by giving free personal consultation. -Otto
Re: locate weirdness
On Thu, Jan 12, 2012 at 03:28:37PM +0100, Fritz Wuehler wrote: > Hey Henning, > > > off-topic diatribes? coming to this mailing list asking for help about > > a 4 year old release when it is clearly documented that you are > > ENTIRELY ON YOUR OWN with releases older than a year is at least off > > topic, if not outright rude. > > I notice you spend much more time scolding people than actually saying > anything worthwhile. You should work on yourself and find out why that > is. Perhaps you could benefit from some anger management training? > > > so my advice is to upgrade. > > You could have said just that without the assholier than thou 'tude. I know, > you can't help it. > Something you and he apparently have in common. henning balances it out by contributing code. What have you contributed? Ken
Re: locate weirdness
On , "LV Lammert" wrote: > Recommendations to upgrade are total BS - the system is 4.3 for reasons > which I will not share with the list because they are not germaine to any > issue raised herein. Such comments (beyond Theo's first one, to which he > is more than entltled) are pure Obsd MISC - off topic, provide no useful > information, and only worth reading for entertainment value. Upgrading is a rule of this list. It cannot get anymore simple than that.
Re: locate weirdness
At 05:59 AM 1/12/2012, you wrote: Is this a new phenomenon? That is, did it use to work earlier and something is now broken? In that case, what did you change? :-) Yes, .. nothing. (I'm not even gonna comment on the rest of this discussion. Ah, yet I just did, didn't I? Oh well, in that case... Of course! Can't resist ! I feel it would have been prudent of you not to insult this community when denied help because of *your* own failure to play by the rules. ExCUSE me? Who is insulting whom? I asked a simple question [forgetting for a moment the propensities of this list], and was totally flamed for trying to solve a problem. We happily participate in the user community, as you, because OBSD is a quality product maintained by good folks that value technology; the fact that USERS like to crap on OTHER users is exactly why many people just turn off and tune out. Recommendations to upgrade are total BS - the system is 4.3 for reasons which I will not share with the list because they are not germaine to any issue raised herein. Such comments (beyond Theo's first one, to which he is more than entltled) are pure Obsd MISC - off topic, provide no useful information, and only worth reading for entertainment value. Not only is it counterproductive for your own problem's sake, all this annoying bickering that flares up from time to time wears on everyones patience, and I for one don't like to see this community wither away because some individuals can't accept that there are a certain set of rules by which to act in order to both get something out of the community and to contribute to it. ExCUSE me again? If you don't like flame wars, why do YOU participate? Enough said - thanks to the folks that actually had some *CONSTRUCTIVE* suggestions, they are the 'wheat' that provide value on this list, as opposed to folks like yourself that are purely 'chaff'. Lee
Re: locate weirdness
Hey Henning, > off-topic diatribes? coming to this mailing list asking for help about > a 4 year old release when it is clearly documented that you are > ENTIRELY ON YOUR OWN with releases older than a year is at least off > topic, if not outright rude. I notice you spend much more time scolding people than actually saying anything worthwhile. You should work on yourself and find out why that is. Perhaps you could benefit from some anger management training? > so my advice is to upgrade. You could have said just that without the assholier than thou 'tude. I know, you can't help it.
Re: locate weirdness
On 2012-01-11 16.26, L. V. Lammert wrote: > Have a 4.3 server with a really weird problem: locate ONLY indexes one > [user file] partition! IOW, no binaries are indexed, nor is /usr/, /var, .. Is this a new phenomenon? That is, did it use to work earlier and something is now broken? In that case, what did you change? :-) Anyway, I have a very vague recollection of a problem in earlier versions of locate, with filenames containing 8-bit characters. I don't recall the details now, but I *think* that it core dumped, leaving the database in- completely generated. That may very well be an exact match to the symptoms you are seeing now, so check your file systems for (new) files containing unusual characters. (I'm not even gonna comment on the rest of this discussion. Ah, yet I just did, didn't I? Oh well, in that case... I feel it would have been prudent of you not to insult this community when denied help because of *your* own failure to play by the rules. Not only is it counterproductive for your own problem's sake, all this annoying bickering that flares up from time to time wears on everyones patience, and I for one don't like to see this community wither away because some individuals can't accept that there are a certain set of rules by which to act in order to both get something out of the community and to contribute to it. People eventually tire of this shit and leave, it's as simple as that. And rarely is it the annoying, non-contributing people that leave first... That said, if I look at my own modest server farm, my oldest OpenBSD box is a 3.8 one. I dare not touch it these days! So I can absolutely understand that there are perfectly valid reasons to keep an old, non- upgraded machine around. But I most certainly won't expect any help from anyone else but me either, if and when it fails. The decision to keep it frozen in time was mine, and I'll be the one suffering the eventual consequences of that decision, not anyone else. It will go belly-up one day, and when it does I'll replace it with a modern server with a current (as in current release) OpenBSD. And THEN I'll upgrade it regularly so I won't fall years behind on maintenance again. Promise. Really.) Regards, /Benny > All filesystems are ffs; > > I deleted /var/db/locate.db and recreated with > /usr/libexec/locate.updatedb more than once; > > locate.rc is stock: > > == > TMPDIR="/var/tmp" > FCODES="/var/db/locate.database" > SEARCHPATHS="/" > PRUNEPATHS="/tmp /var/tmp /usr/tmp" > FILESYSTEMS="ffs ufs ext2fs" > > [comments pruned] > = > > The locate database seems to be normal: > > Database: /var/db/locate.database > Compression: Front: 19.48%, Bigram: 65.90%, Total: 14.52% > Filenames: 218512, Characters: 14825215, Database size: 2153551 > Bigram characters: 734303, Integers: 5440, 8-Bit characters: 3 > > = > > Trying to troubleshoot a Perl module problem, and locate is not > returning any hits except user files ! > Any pointers on what is 'intefering' with the process? -- 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: locate weirdness
On Wed, 11 Jan 2012 13:47:48 -0600 "L. V. Lammert" wrote: > At 01:30 PM 1/11/2012, Jeremy O'Brien wrote: > > >4.3 was released May 1, 2008. That's almost 4 years old software. What > >are you expecting here? Someone to check out the code from that > >version and deeply inspect what may be causing your problem, that is > >more than likely already fixed in a later version? > > Another typical reply - the question was "has anyone ever seen > anything like this", .. or, perhaps, "what could be causing it". No > need for the off-topic diatribes - a simple no would more than suffice. > > Lee > Yes. I have seen problems with locate in all releases that I have run. Unfortunately I have never bothered to characterize these problems. Dhu
Re: locate weirdness
FWIW, on my system it seems to work (and yes, I should upgrade!) $ uname -a OpenBSD foo.bar 4.3 GENERIC.MP#587 i386 On Wed, Jan 11, 2012 at 03:09:50PM -0600, L. V. Lammert wrote: > On Wed, 11 Jan 2012, Philip Guenther wrote: > > > Lesson #1: examine the anomalous data for clues. > > > > So, you're saying that > > locate /usr | grep ^/usr | head $ locate /usr | grep '^/usr' | head /usr ... > > returns nothing but > > > Yep! As does locate /usr > > > locate /home | grep ^/home | head $ locate /home | grep '^/home' | head /home ... > > returns something? (/home being a stand-in for whatever your unsaid > > "[user file] partition" is) > > > > Perhaps you should investigate how those two directories differ? > > > That was the original question - both are ffs, both are rw, the only > difference between then that /home is nosuid, however that does not > affect locate on 3.3, 4.9, or 5.0 (just tested). $ mount /dev/sd0a on / type ffs (local, with quotas) /dev/sd1a on /home type ffs (local, nodev, nosuid, with quotas) /dev/sd2a on /var type ffs (local, nodev, nosuid, with quotas) > TFTR! > > Lee -- Martin Bock :wq
Re: locate weirdness
* L. V. Lammert [2012-01-11 20:48]: > Another typical reply - the question was "has anyone ever seen > anything like this", .. or, perhaps, "what could be causing it". No > need for the off-topic diatribes - a simple no would more than > suffice. ah, you think the other kids won't find you when you cover your eyes. off-topic diatribes? coming to this mailing list asking for help about a 4 year old release when it is clearly documented that you are ENTIRELY ON YOUR OWN with releases older than a year is at least off topic, if not outright rude. i have extremely vague memories of something weird with locate somewhen in the past. might misremember, doesn't matter, haven't seen anything like that for a long time, so my advice is to upgrade. -- 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, Root to Fully Managed Henning Brauer Consulting, http://henningbrauer.com/
Re: locate weirdness
On Jan 11, 2012, at 4:08 PM, L. V. Lammert wrote: > On Wed, 11 Jan 2012, Philip Guenther wrote: > >>> Agreed, .. but if locate.update does NOT run as root, that would seem to >>> indicate some problem other than permissions. >> >> If you're saying what I think you're saying, then I disagree and think >> your logic is backwards. >> What user do you think locate.updatedb is run as? >> > If it does not run as root, then it isn't a permission issue as running as > root provides all required permissions, eh? eh? "if it does not run as root. . . running as root provides. . ."?? To put it bluntly, if updatedb runs as root, it has all possible permissions. If updatedb does NOT run as root, it does NOT have all possible permissions. > I have never seen locate.updatedb fail when run as root (3.0 to 5.0, > actually), .. but, then, it isn't exactly 'failing', it just isn't > indexing anything except "/home". FWIW, I've never had a problem with locate since, oh, I think 2.6. But the point is, *IS* updatedb running as root? > The only other possible hypothesis, is that it is running out of memory; > one would expect some sort of error to be returned in that case and a > blank database as a result, not one partially populated. No, your logic is backward, as Philip has been gently pointing out. So, to diagnose your problem (regardless of release -- this is diagnosing 101 here): 1) Find out *EXACTLY* how updatedb is being called, and run it, except don't redirect errors to /dev/null or files or such. Check for error messages and/or exit codes 2) Since updatedb is a *SHELL SCRIPT*, try running it with -x (this breaks 1), of course). If the above is not enough for you to figure it out, email me off list and I'll help. But I don't have a 4.3 machine handy (I have a 4.6 and a 4.7 machine). Sean [demime 1.01d removed an attachment of type application/pkcs7-signature which had a name of smime.p7s]
Re: locate weirdness
On Wed, Jan 11, 2012 at 4:08 PM, L. V. Lammert wrote: > On Wed, 11 Jan 2012, Philip Guenther wrote: > >> > Agreed, .. but if locate.update does NOT run as root, that would seem to >> > indicate some problem other than permissions. >> >> If you're saying what I think you're saying, then I disagree and think >> your logic is backwards. >> What user do you think locate.updatedb is run as? >> > If it does not run as root, then it isn't a permission issue as running as > root provides all required permissions, eh? I'm sorry, but I don't understand that sentence. It appears to conflate running as root with not running as root, or I'm miscounting the 'not's. So let me try again: what user do you think locate.updatedb is run as? > I have never seen locate.updatedb fail when run as root (3.0 to 5.0, > actually), .. but, then, it isn't exactly 'failing', it just isn't > indexing anything except "/home". I don't understand this sentence either. If the word "fail" is ambiguous or unclear, then use a different word instead! I cannot tell from what you wrote what behavior you saw when you manually ran locate.updatedb as root on (say) 3.0. Philip Guenther
Re: locate weirdness
On 01/11/12 14:24, Barry Grumbine wrote: >> Bite the bullet, upgrade, life is better at 5.0 >> > > ...knew I forgot something. > > There aren't many North American mirrors that go back to 4.2. I was > fortunate to find "obsd.cec.mtu.edu" which Nick Holland recently > notified us that he needs to take down very soon. Ouch. I opted to archive old versions of OpenBSD for historical interest... When did we get Mozilla ported? What platforms were supported back in the 2.5 days? What was it like to install OpenBSD 2.0 on a 386 system? Etc. It's FUN. And, storage is cheap; for something like $500US several years ago, I was able to add 1.5G of redundant storage to obsd.cec.mtu.edu, and that allowed me to make a comprehensive archive available, and being it was my money and my interest, I did. :) Same reason I collect 80+ year old calculating devices and 40 year old calculators...but this, I can easily share with others. It was certainly never intended to be USED for production. It bothers me that people may have been using my archive to avoid upgrading ('specially since I write the upgrade guides!). (for anyone tempted to snarf down all my old archival versions of OpenBSD before the final shut down of obsd.cec.mtu.edu, don't worry, I believe I'll be able to get all the hardware (20U worth! loaded with data) back, so the data won't be vanishing into thin air. And, I don't believe it is the only copy left on the 'net.) I've removed everything from 4.8 and before from the "easily spotted" space on the mirror, so it is now only available in a directory clearly marked "archive". Nick.
Re: locate weirdness
On Wed, 11 Jan 2012, Philip Guenther wrote: > > Agreed, .. but if locate.update does NOT run as root, that would seem to > > indicate some problem other than permissions. > > If you're saying what I think you're saying, then I disagree and think > your logic is backwards. > What user do you think locate.updatedb is run as? > If it does not run as root, then it isn't a permission issue as running as root provides all required permissions, eh? I have never seen locate.updatedb fail when run as root (3.0 to 5.0, actually), .. but, then, it isn't exactly 'failing', it just isn't indexing anything except "/home". The only other possible hypothesis, is that it is running out of memory; one would expect some sort of error to be returned in that case and a blank database as a result, not one partially populated. Lee
Re: locate weirdness
On Wed, Jan 11, 2012 at 3:18 PM, L. V. Lammert wrote: > On Wed, 11 Jan 2012, Philip Guenther wrote: ... >> Ah, but that's *not* how locate.updatedb is invoked by the cronjob! >> There's a reason I called out the need to mimic that when trying to >> replicate the problem while walking through locate.updatedb >> manually... >> > Agreed, .. but if locate.update does NOT run as root, that would seem to > indicate some problem other than permissions. If you're saying what I think you're saying, then I disagree and think your logic is backwards. What user do you think locate.updatedb is run as? Philip Guenther
Re: locate weirdness
Am 12.01.12 00:13, schrieb Philip Guenther: On Wed, Jan 11, 2012 at 3:02 PM, Marian Hettwer wrote: ... ([foobar@bistromath]<~>)$ time sudo /usr/libexec/locate.updatedb Password: Ah, but that's *not* how locate.updatedb is invoked by the cronjob! There's a reason I called out the need to mimic that when trying to replicate the problem while walking through locate.updatedb manually... [root@bistromath] <~> # /bin/sh /etc/weekly Rebuilding locate database: Rebuilding whatis databases: [root@bistromath] <~> # echo $? 0 still on OpenBSD 4.0. And /etc/weekly looks like a reasonable easy straight forward shell script. (I would expect nothing else in OpenBSD). ./Marian
Re: locate weirdness
On Wed, 11 Jan 2012, Philip Guenther wrote: > On Wed, Jan 11, 2012 at 3:02 PM, Marian Hettwer wrote: > ... > > ([foobar@bistromath] <~>)$ time sudo /usr/libexec/locate.updatedb > > Password: > > Ah, but that's *not* how locate.updatedb is invoked by the cronjob! > There's a reason I called out the need to mimic that when trying to > replicate the problem while walking through locate.updatedb > manually... > Agreed, .. but if locate.update does NOT run as root, that would seem to indicate some problem other than permissions. BTW - Looked at a couple of other possiblities, .. mysql had a lot of space in log files so I freed up most of them, no change; the other possibility could be that of a memory problem, but I have no knowledge of 'bigmem' and how that works. Lee
Re: locate weirdness
On Wed, Jan 11, 2012 at 3:02 PM, Marian Hettwer wrote: ... > ([foobar@bistromath] <~>)$ time sudo /usr/libexec/locate.updatedb > Password: Ah, but that's *not* how locate.updatedb is invoked by the cronjob! There's a reason I called out the need to mimic that when trying to replicate the problem while walking through locate.updatedb manually... Philip Guenther
Re: locate weirdness
Am 11.01.12 22:34, schrieb Ted Unangst: On Wed, Jan 11, 2012, L. V. Lammert wrote: At 01:30 PM 1/11/2012, Jeremy O'Brien wrote: 4.3 was released May 1, 2008. That's almost 4 years old software. What are you expecting here? Someone to check out the code from that version and deeply inspect what may be causing your problem, that is more than likely already fixed in a later version? Another typical reply - the question was "has anyone ever seen anything like this", .. or, perhaps, "what could be causing it". No need for the off-topic diatribes - a simple no would more than suffice. okie, dokie. locate works for me! Ah! History Channel. /me too haz workin locate ([foobar@bistromath] <~>)$ locate pfctl /sbin/pfctl /usr/sbin/ospfctl ([foobar@bistromath] <~>)$ uname -a OpenBSD bistromath.meganet.local 4.0 GENERIC#1107 i386 ([foobar@bistromath] <~>)$ time sudo /usr/libexec/locate.updatedb Password: real0m9.379s user0m1.453s sys 0m3.406s ([foobar@bistromath] <~>)$ echo $? 0 I really should update this system ;-) ./Marian
Re: locate weirdness
On Wed, Jan 11, 2012, L. V. Lammert wrote: > At 01:30 PM 1/11/2012, Jeremy O'Brien wrote: > >>4.3 was released May 1, 2008. That's almost 4 years old software. What >>are you expecting here? Someone to check out the code from that >>version and deeply inspect what may be causing your problem, that is >>more than likely already fixed in a later version? > > Another typical reply - the question was "has anyone ever seen > anything like this", .. or, perhaps, "what could be causing it". No > need for the off-topic diatribes - a simple no would more than suffice. okie, dokie. locate works for me!
Re: locate weirdness
L. V. Lammert wrote: > At 01:04 PM 1/11/2012, Barry Grumbine wrote: >>Bite the bullet, upgrade, life is better at 5.0 > > Sorry, but *UPGRADING* isn't the question - the question is why > locate is not working properly. If nobody has ever seen such a > problem, it would be quite more forthright to just admit that than > spout the normal crap this list promulgates. But, then, I should have > expected multiple replies that are off topic, of no help, and not > worth the time to read. Sorry, I had momentarily forgotten the > definition of OBSD Misc - my bad. > > If nobody can answer the question, that's is not a problem, just say so! > > Lee > > Why don't you download 5.0 on a separate disc or folder and then compare the differences using a diff tool to see what changed and if it is fixed in 5.0 you can apply a patch to your old one. find the problem files, compare them to the new code. Also make sure the bug isn't in the new release because if it is, you should report it.
Re: locate weirdness
On Wed, Jan 11, 2012 at 1:09 PM, L. V. Lammert wrote: > On Wed, 11 Jan 2012, Philip Guenther wrote: >> Lesson #1: examine the anomalous data for clues. >> >> So, you're saying that >> locate /usr | grep ^/usr | head >> >> returns nothing but > > Yep! As does locate /usr > >> locate /home | grep ^/home | head >> >> returns something? (/home being a stand-in for whatever your unsaid >> "[user file] partition" is) >> >> Perhaps you should investigate how those two directories differ? >> > That was the original question - both are ffs, both are rw, the only > difference between then that /home is nosuid, however that does not > affect locate on 3.3, 4.9, or 5.0 (just tested). If you've established that the two directories have no differences in mode, etc, then I guess you'll have to go with the "walk through things step by step" path then. Philip Guenther
Re: locate weirdness
Also, in order to help others when they encounter a similar issue, please be sure to post what the problem and/or solution were once you figure them out. Philip Guenther
Re: locate weirdness
On Wed, 11 Jan 2012, Philip Guenther wrote: > Lesson #1: examine the anomalous data for clues. > > So, you're saying that > locate /usr | grep ^/usr | head > > returns nothing but > Yep! As does locate /usr > locate /home | grep ^/home | head > > returns something? (/home being a stand-in for whatever your unsaid > "[user file] partition" is) > > Perhaps you should investigate how those two directories differ? > That was the original question - both are ffs, both are rw, the only difference between then that /home is nosuid, however that does not affect locate on 3.3, 4.9, or 5.0 (just tested). TFTR! Lee
Re: locate weirdness
On Wed, 11 Jan 2012, Philip Guenther wrote: > Also, in order to help others when they encounter a similar issue, > please be sure to post what the problem and/or solution were once you > figure them out. > > Philip Guenther > Amen! At least there's a chance it would turn up in the search engines. Lee
Re: locate weirdness
Time for today's "how to debug a problem" lesson. On Wed, Jan 11, 2012 at 7:26 AM, L. V. Lammert wrote: > Have a 4.3 server with a really weird problem: locate ONLY indexes one [user > file] partition! IOW, no binaries are indexed, nor is /usr/, /var, .. Lesson #1: examine the anomalous data for clues. So, you're saying that locate /usr | grep ^/usr | head returns nothing but locate /home | grep ^/home | head returns something? (/home being a stand-in for whatever your unsaid "[user file] partition" is) Perhaps you should investigate how those two directories differ? > The locate database seems to be normal: > > Database: /var/db/locate.database > Compression: Front: 19.48%, Bigram: 65.90%, Total: 14.52% > Filenames: 218512, Characters: 14825215, Database size: 2153551 > Bigram characters: 734303, Integers: 5440, 8-Bit characters: 3 Lesson #2: step through the problem computation and verify the correctness of intermediate stages. So you've run locate.updatedb manually. It's just a shell script, so perhaps you should run the commands in it manually, one by one and examining the intermediate output of pipes, etc. Be sure to do so in a shell that reproduces how locate.updatedb is called from /etc/weekly! Philip Guenther
Re: locate weirdness
On Wed, 11 Jan 2012, Marian Hettwer wrote: > Hi, > > > Am 11.01.12 20:17, schrieb L. V. Lammert: > > At 01:04 PM 1/11/2012, Barry Grumbine wrote: > >> Bite the bullet, upgrade, life is better at 5.0 > > > > Sorry, but *UPGRADING* isn't the question - the question is why locate > > is not working properly. > > No. You were advised to upgrade, since 4.3 is not supported anymore. > Heck, probably nobody can even remember whether something was odd with > locate in 4.3. > Upgrade to a supported release and if you still face problems, come back > to the list. > > Try to look from a different angle here. > Say, you would have an old Debian Sarge release (years old) and you > would approach a debian mailing list with "something is weird with > locate", pretty sure you would get a lot of advises to upgrade first, > test then, and if problem persists, come back. > > All good and jolly! > > ./Marian > Hope you got off on the bs, .. as usual, offtopic, nothing useful, not worth readying. Quite repetitive of the other BS, actually. Lee
Re: locate weirdness
Hi, Am 11.01.12 20:17, schrieb L. V. Lammert: At 01:04 PM 1/11/2012, Barry Grumbine wrote: Bite the bullet, upgrade, life is better at 5.0 Sorry, but *UPGRADING* isn't the question - the question is why locate is not working properly. No. You were advised to upgrade, since 4.3 is not supported anymore. Heck, probably nobody can even remember whether something was odd with locate in 4.3. Upgrade to a supported release and if you still face problems, come back to the list. Try to look from a different angle here. Say, you would have an old Debian Sarge release (years old) and you would approach a debian mailing list with "something is weird with locate", pretty sure you would get a lot of advises to upgrade first, test then, and if problem persists, come back. All good and jolly! ./Marian
Re: locate weirdness
On Wed, Jan 11, 2012 at 14:47, L. V. Lammert wrote: > At 01:30 PM 1/11/2012, Jeremy O'Brien wrote: > >> 4.3 was released May 1, 2008. That's almost 4 years old software. What >> are you expecting here? Someone to check out the code from that >> version and deeply inspect what may be causing your problem, that is >> more than likely already fixed in a later version? > > > Another typical reply - the question was "has anyone ever seen anything like > this", .. or, perhaps, "what could be causing it". No need for the off-topic > diatribes - a simple no would more than suffice. > OK then. I have used OpenBSD since 4.0, and I have not seen this behavior. I recommend seeing if an upgrade fixes your problem. ;)
Re: locate weirdness
> They were valid replies, but a straw man argument at best. I think he > would have preferred to hear something more like: > > "Yeah, I saw something similar happen on my systems running an older > release. I don't really remember the release, but I do remember the > problem eventually went away for me. I'm not really sure of what's > happening, but my best guess is giving the latest release a try and > seeing if that solves the problem for you." The OP has been around long enough to know we don't like talking about ancient code. It is completely FAQ, and he knows better. It's not a kernel crash. It's not pf letting packets through. It's locate. Come on.
Re: locate weirdness
At 01:30 PM 1/11/2012, Jeremy O'Brien wrote: 4.3 was released May 1, 2008. That's almost 4 years old software. What are you expecting here? Someone to check out the code from that version and deeply inspect what may be causing your problem, that is more than likely already fixed in a later version? Another typical reply - the question was "has anyone ever seen anything like this", .. or, perhaps, "what could be causing it". No need for the off-topic diatribes - a simple no would more than suffice. Lee
Re: locate weirdness
On 01/11/2012 02:30 PM, Jeremy O'Brien wrote: On Wed, Jan 11, 2012 at 14:17, L. V. Lammert wrote: At 01:04 PM 1/11/2012, Barry Grumbine wrote: Bite the bullet, upgrade, life is better at 5.0 Sorry, but *UPGRADING* isn't the question - the question is why locate is not working properly. If nobody has ever seen such a problem, it would be quite more forthright to just admit that than spout the normal crap this list promulgates. But, then, I should have expected multiple replies that are off topic, of no help, and not worth the time to read. Sorry, I had momentarily forgotten the definition of OBSD Misc - my bad. If nobody can answer the question, that's is not a problem, just say so! Lee 4.3 was released May 1, 2008. That's almost 4 years old software. What are you expecting here? Someone to check out the code from that version and deeply inspect what may be causing your problem, that is more than likely already fixed in a later version? The replies were perfectly valid and helpful. In the software world, you're using an antique. They were valid replies, but a straw man argument at best. I think he would have preferred to hear something more like: "Yeah, I saw something similar happen on my systems running an older release. I don't really remember the release, but I do remember the problem eventually went away for me. I'm not really sure of what's happening, but my best guess is giving the latest release a try and seeing if that solves the problem for you." -- David Cantrell WH6DSN | http://blog.burdell.org/
Re: locate weirdness
On Wed, Jan 11, 2012 at 14:17, L. V. Lammert wrote: > At 01:04 PM 1/11/2012, Barry Grumbine wrote: >> >> Bite the bullet, upgrade, life is better at 5.0 > > > Sorry, but *UPGRADING* isn't the question - the question is why locate is > not working properly. If nobody has ever seen such a problem, it would be > quite more forthright to just admit that than spout the normal crap this > list promulgates. But, then, I should have expected multiple replies that > are off topic, of no help, and not worth the time to read. Sorry, I had > momentarily forgotten the definition of OBSD Misc - my bad. > > If nobody can answer the question, that's is not a problem, just say so! > >Lee > 4.3 was released May 1, 2008. That's almost 4 years old software. What are you expecting here? Someone to check out the code from that version and deeply inspect what may be causing your problem, that is more than likely already fixed in a later version? The replies were perfectly valid and helpful. In the software world, you're using an antique.
Re: locate weirdness
> Bite the bullet, upgrade, life is better at 5.0 > ...knew I forgot something. There aren't many North American mirrors that go back to 4.2. I was fortunate to find "obsd.cec.mtu.edu" which Nick Holland recently notified us that he needs to take down very soon. After Looking through all the mirrors, I think the only OpenBSD archive located in North America is planetunix.net : ftp://mirror.planetunix.net/pub/OpenBSD/4.3/ Most of the rest only mirror a couple of releases. -Barry
Re: locate weirdness
At 01:04 PM 1/11/2012, Barry Grumbine wrote: Bite the bullet, upgrade, life is better at 5.0 Sorry, but *UPGRADING* isn't the question - the question is why locate is not working properly. If nobody has ever seen such a problem, it would be quite more forthright to just admit that than spout the normal crap this list promulgates. But, then, I should have expected multiple replies that are off topic, of no help, and not worth the time to read. Sorry, I had momentarily forgotten the definition of OBSD Misc - my bad. If nobody can answer the question, that's is not a problem, just say so! Lee
Re: locate weirdness
On Wed, Jan 11, 2012 at 9:49 AM, L. V. Lammert wrote: > At 10:41 AM 1/11/2012, Theo de Raadt wrote: >> >> > Have a 4.3 server [rest deleted] >> >> There is a ton of documentation that makes it clear you are on your >> own more than two releases back. > > > So, you're advocating incomplete information? Is that not a bigger problem? > >Lee Bite the bullet, upgrade, life is better at 5.0 Prior to last year, I had been upgrading OpenBSD by clean install, then install and configure all packages, it was a major hassle. Things hit critical mass about a year ago. I had five systems on either 4.2 or 4.3. I gave in and decided to learn the "new" upgrade tools. As a result I upgraded nearly all my 10+ OpenBSD systems to 4.9 over the course of 3-4 weeks. That was a total of 43 upgrades counting each release on each system. I'm more confident in my OpenBSD systems now and the 4.9->5.0 upgrades went even more smoothly (once I figured out /etc/rc.d/). I'm not sure how I missed them before, but follow the upgrade guides is essential: http://www.openbsd.org/faq/upgrade50.html Once you get to 4.3, upgrading your systems becomes a whole lot easier, thanks in large part to sysmerge(8), introduced in 4.4 The upgrade from 4.2 -> 4.3 was still a bit of a hassle, but after that upgrades went fairly smooth with little incident. It is very interesting to see sysmerge(8) get better and better with each release. Go get'em man, those upgrades are nowhere near as hard as they once were, back in the day when you had to grep 10,000 LOC, uphill, both ways, just to get schooled. Have a nice day, Barry
Re: locate weirdness
> So, you're advocating incomplete information? Is that not a bigger problem? No, we don't support old releases. 4.3 is very old. You should update your OS to something supported, and likely your problem will go away.
Re: locate weirdness
>> > Have a 4.3 server [rest deleted] >> >>There is a ton of documentation that makes it clear you are on your >>own more than two releases back. > >So, you're advocating incomplete information? Is that not a bigger problem? No, I am advocating that you TAKE CARE OF YOUR OWN PROBLEMS YOURSELF. We do not support old releases. AT ALL.
Re: locate weirdness
At 10:41 AM 1/11/2012, Theo de Raadt wrote: > Have a 4.3 server [rest deleted] There is a ton of documentation that makes it clear you are on your own more than two releases back. So, you're advocating incomplete information? Is that not a bigger problem? Lee
Re: locate weirdness
> Have a 4.3 server [rest deleted] There is a ton of documentation that makes it clear you are on your own more than two releases back.
locate weirdness
Have a 4.3 server with a really weird problem: locate ONLY indexes one [user file] partition! IOW, no binaries are indexed, nor is /usr/, /var, .. All filesystems are ffs; I deleted /var/db/locate.db and recreated with /usr/libexec/locate.updatedb more than once; locate.rc is stock: == TMPDIR="/var/tmp" FCODES="/var/db/locate.database" SEARCHPATHS="/" PRUNEPATHS="/tmp /var/tmp /usr/tmp" FILESYSTEMS="ffs ufs ext2fs" [comments pruned] = The locate database seems to be normal: Database: /var/db/locate.database Compression: Front: 19.48%, Bigram: 65.90%, Total: 14.52% Filenames: 218512, Characters: 14825215, Database size: 2153551 Bigram characters: 734303, Integers: 5440, 8-Bit characters: 3 = Trying to troubleshoot a Perl module problem, and locate is not returning any hits except user files ! Any pointers on what is 'intefering' with the process? Lee