Re: informations about boot sequence (Re: F15 - mysql start problem)
W dniu 10 maja 2011 01:23 użytkownik Stephen John Smoogen
napisał:
> 2011/5/9 Michał Piotrowski
>>
>> 2011/5/10 Lennart Poettering :
>> > On Mon, 09.05.11 23:54, Michał Piotrowski (mkkp...@gmail.com) wrote:
>> >> No, only for /run/user/ - because there is a simple workaround that
>> >> can be used on affected systems if the administrator considers his
>> >> system as vulnerable for malicious users.
>> >
>> > Again, we had /dev/shm for years on Linux. This weakness in the security
>> > model is not news, not at all.
>>
>> Yes, but /run/user is a new thing and it gives wonderful opportunity
>> to DoS services for all system users. Thats my POV. And my POV is that
>> it should be documented - users should be aware about this. Also FPL
>> agreed with my arguments.
>>
>> That's all. If you do not agree with that, then I'm giving up :)
>>
>
> Let's make this simple:
>
> FAQ: How can I make my system unusable? How can I create a denial of service?
>
> Answer: On default systems there are multiple ways to do this, please
> choose one or more of the following:
>
> a) Denial of CPU. The Fork Bomb is the standard way to kill a system:
> In a shell type the following:
>
> :(){ :|: & };:
>
> perl -e 'fork while fork'
I wonder if there is a way to add some memory/cpu time/etc restrictions to
/sys/fs/cgroup/systemd/user/
control groups using systemd. systemd already isolates the user
processes in control groups, so adding the ability to add restrictions
is probably not a bad idea from my POV.
>
> b) Denial of Filesystem. There are several ways of doing this. Usually
> it can be done quickly by the following:
>
> dd if=/dev/zero of=
>
> is a standard, but easily fixed by deleting one file. Adding some
> flare you can great randomly created files in multiple places.
>
> Places of entry where a system can cause problems are the following:
>
> /tmp/
> /var/tmp/
> /dev/shmem/
> /run/file/
>
> c) Denial of Logs
> while true; do
> logger $( dd if=/dev/urandom count=1 bs=128 2> /dev/null |tr -dC
> '[:print:]' )
> done
>
> d) Denial of service via audits
> while true; do
> cat /etc/shadow
> done
>
>
> Doing a, b, c, and d at the same time is always fun for the family.
> There are many other ways you as a user can cause problems to your own
> system...
>
> --
> Stephen J Smoogen.
> "The core skill of innovators is error recovery, not failure avoidance."
> Randy Nelson, President of Pixar University.
> "Let us be kind, one to another, for most of us are fighting a hard
> battle." -- Ian MacLaren
> --
> devel mailing list
> devel@lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/devel
>
--
Best regards,
Michal
http://eventhorizon.pl/
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Re: anaconda-15.31-1.fc15 - Update the requirements for memory..
On Tue, 2011-05-10 at 11:11 +0900, Misha Shnurapet wrote: > 10.05.2011, 05:33, "Adam Williamson" : > > this is already known and discussed at: > > 768 megs for an installer? Sweet Georgia Brown of Kingston town, I think 512 > is pretty much. The links explain the situation. > > https://bugzilla.redhat.com/show_bug.cgi?id=680542 > > Not specifically related, just asks for a correct message error. Yeah, but it's one of the various places where the issue is discussed. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: Is Rawhide supposed to be useful?
On Tue, May 10, 2011 at 11:18:54AM +1000, Peter Hutterer wrote: > This is my fault, sorry. I updated the server but missed out on rebuilding > the drivers. And with one thing leading to another, Easter came, I forgot > about it and the above bug didn't show up on my radar until ajax pinged me > this morning. Thank you very much. -- Matthew Miller Senior Systems Architect -- Instructional & Research Computing Services Harvard School of Engineering & Applied Sciences -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: anaconda-15.31-1.fc15 - Update the requirements for memory..
10.05.2011, 05:33, "Adam Williamson" : > this is already known and discussed at: 768 megs for an installer? Sweet Georgia Brown of Kingston town, I think 512 is pretty much. > https://bugzilla.redhat.com/show_bug.cgi?id=680542 Not specifically related, just asks for a correct message error. -- Best regards, Misha Shnurapet, Fedora Project Contributor https://fedoraproject.org/wiki/Shnurapet shnurapet AT fedoraproject.org, GPG: 00217306 -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: Is Rawhide supposed to be useful?
On Tue, 2011-05-10 at 11:18 +1000, Peter Hutterer wrote: > I'm about to rebuild and update the drivers, but my upstream link is > currently using pigeons, so it will take me a while. I tried that, but I couldn't stuff the squawking things into the HDMI port... -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: Interlinux waiting list
On Mon, 2011-05-09 at 21:14 -0300, Sergio Belkin wrote: > Sorry... Did I miss something? I'd imagine someone signed an interlinux address up to the list, and that was an automated response. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Miloslav Trmač : > 2011/5/10 Stephen John Smoogen : >> Let's make this simple: >> >> FAQ: How can I make my system unusable? How can I create a denial of service? >> >> Answer: On default systems there are multiple ways to do this, please >> choose one or more of the following: > > That's all true, on the other hand there are countermeasures > available; in larger organizations the countermeasures are documented, > configured on each system, and their presence is periodically > verified. 1) Running Fedora on large organizations is usually heavily frowned on. Been there, fought that battle, realized it was a lost issue, moved on. If Fedora is run, it always seemed to be more sandboxed away from production than all the more vulnerable systems. 2) The number of large organizations where RHEL, SuSE, etc where it is documented is very high. The number where it is configured is no where near as high. The place where those configs are turned off because some sysadmin hates being told how to set up his stuff is probably just as high. 3) The number of Unix sysadmins who I have met who work for such large organizations who HAVE no idea about any of those problems is also very high. I have gone through too many firedrills about one or more of the local DOS's because some guy who had been a sysadmin for 10+ years all of a sudden discovered his bullet proof Unix system can be brought down by it. And then I spent a long time finding out that all of the production software depended on something that the fix cleaned up. 4) The fixes to any of those problems are usable but usually end up breaking something else. Why did we have to turn off fork bomb protection? Because "big named business software package" broke because it was enabled. Make /var/tmp and /tmp separate partitions.. oh look another "big named software package uses hard links with its software in /opt to /tmp and won't work." Put mount options on /dev/shm and poof "all that web stuff this server is supposed to run doesn't." 5) The /dev/shm problem (and now /run/files) has been around for a LONG time. I have been finding root kits there since 2002 (and remember it being mentioned a lot earlier). Making it so it wasn't world writable but say group writable never seemed to fix the issues because the exploitable programs always needed to be in the groups to use it. The real world rule of security is "Unless a problem is fixed out of the box, it is not going to be fixed." Worrying about /run/file when we "ship" worse problems out of the gate is another closing the barn door after all the horses have run away. Especially when a lot of the issues were going to be fixed some time or another out of the past. I am not saying we shouldn't document it, and I am not saying we shouldn't fix it now but we had better do a darn better job with all the rest of the problems we come with. -- Stephen J Smoogen. "The core skill of innovators is error recovery, not failure avoidance." Randy Nelson, President of Pixar University. "Let us be kind, one to another, for most of us are fighting a hard battle." -- Ian MacLaren -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: Is Rawhide supposed to be useful?
On Mon, May 09, 2011 at 08:49:28AM -0400, Matthew Miller wrote: > https://bugzilla.redhat.com/show_bug.cgi?id=698237 > > and it's been this way for a while: > > http://article.gmane.org/gmane.linux.redhat.fedora.devel/147699 > > which just says > > Rawhide's X is going to be like that for a while. We have ABI version > checks in place, but they only operate off the numbers actually exported > from the server, and those don't change until ABI freeze for the server > release. > > - ajax > > which is not very helpful -- and that was several weeks ago. The bug above > keeps accumlating duplicates. > > Can some resources be put into making sure this is resolved and protected > against via RPM dependencies? This is my fault, sorry. I updated the server but missed out on rebuilding the drivers. And with one thing leading to another, Easter came, I forgot about it and the above bug didn't show up on my radar until ajax pinged me this morning. I'm about to rebuild and update the drivers, but my upstream link is currently using pigeons, so it will take me a while. The resources have been put in, there are ABI checks in the RPMs but as ajax said, they don't work until the ABI is bumped upstream. Cheers, Peter -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, May 10, 2011 at 1:33 AM, Lennart Poettering wrote: > On Tue, 10.05.11 01:31, Miloslav Trmač (m...@volny.cz) wrote: > >> >> 2011/5/10 Stephen John Smoogen : >> > Let's make this simple: >> > >> > FAQ: How can I make my system unusable? How can I create a denial of >> > service? >> > >> > Answer: On default systems there are multiple ways to do this, please >> > choose one or more of the following: >> >> That's all true, on the other hand there are countermeasures >> available; in larger organizations the countermeasures are documented, >> configured on each system, and their presence is periodically >> verified. > > Countermeasures for the /dev/shm issue? I don't know of any. tmpfs > doesn't do quota. That's the key problem here. mount options, file permissions, SELinux. Perhaps not something that you'd want to do on a general-purpose desktop, but quite reasonable for a single-purpose server. Mirek -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: Interlinux waiting list
Sorry... Did I miss something? -- -- Sergio Belkin http://www.sergiobelkin.com Watch More TV http://sebelk.blogspot.com LPIC-2 Certified - http://www.lpi.org -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On 05/09/2011 08:38 PM, Lennart Poettering wrote: > a) people who manually change the IP address fo mysql to bind on > specific ip addresses, manually also enable > NetworkManager-wait-online.service. This proposal makes the assumption and depends on users actually using NetworkManager as their networking handling application which might not be as close to reality in the server world as you might think even for arguments sake we would say that we default to using NetworkManager, upstream needs to have a solid solution that works across all distro's and what ever network application downstream might be using. In addition to the above any arguments surrounding "boot time" or dynamic roaming network of somekind are also irrelvent in this case since I have yet to meet the sysadmin that a) cares about shaving few seconds of his blades or bricks at boot up or b) picks them up and plugs them in another datacenter.. Which leaves b) and c) as points for discussion. JBG -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, 10.05.11 01:31, Miloslav Trmač (m...@volny.cz) wrote: > > 2011/5/10 Stephen John Smoogen : > > Let's make this simple: > > > > FAQ: How can I make my system unusable? How can I create a denial of > > service? > > > > Answer: On default systems there are multiple ways to do this, please > > choose one or more of the following: > > That's all true, on the other hand there are countermeasures > available; in larger organizations the countermeasures are documented, > configured on each system, and their presence is periodically > verified. Countermeasures for the /dev/shm issue? I don't know of any. tmpfs doesn't do quota. That's the key problem here. Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/10 Stephen John Smoogen : > Let's make this simple: > > FAQ: How can I make my system unusable? How can I create a denial of service? > > Answer: On default systems there are multiple ways to do this, please > choose one or more of the following: That's all true, on the other hand there are countermeasures available; in larger organizations the countermeasures are documented, configured on each system, and their presence is periodically verified. Adding new ways to use the same basic vulnerability is not a "big deal", at the same time there is a class of users who will want to know about this to be able to update their configuration. A release note is the least we can do. Mirek -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Michał Piotrowski
>
> 2011/5/10 Lennart Poettering :
> > On Mon, 09.05.11 23:54, Michał Piotrowski (mkkp...@gmail.com) wrote:
> >> No, only for /run/user/ - because there is a simple workaround that
> >> can be used on affected systems if the administrator considers his
> >> system as vulnerable for malicious users.
> >
> > Again, we had /dev/shm for years on Linux. This weakness in the security
> > model is not news, not at all.
>
> Yes, but /run/user is a new thing and it gives wonderful opportunity
> to DoS services for all system users. Thats my POV. And my POV is that
> it should be documented - users should be aware about this. Also FPL
> agreed with my arguments.
>
> That's all. If you do not agree with that, then I'm giving up :)
>
Let's make this simple:
FAQ: How can I make my system unusable? How can I create a denial of service?
Answer: On default systems there are multiple ways to do this, please
choose one or more of the following:
a) Denial of CPU. The Fork Bomb is the standard way to kill a system:
In a shell type the following:
:(){ :|: & };:
perl -e 'fork while fork'
b) Denial of Filesystem. There are several ways of doing this. Usually
it can be done quickly by the following:
dd if=/dev/zero of=
is a standard, but easily fixed by deleting one file. Adding some
flare you can great randomly created files in multiple places.
Places of entry where a system can cause problems are the following:
/tmp/
/var/tmp/
/dev/shmem/
/run/file/
c) Denial of Logs
while true; do
logger $( dd if=/dev/urandom count=1 bs=128 2> /dev/null |tr -dC
'[:print:]' )
done
d) Denial of service via audits
while true; do
cat /etc/shadow
done
Doing a, b, c, and d at the same time is always fun for the family.
There are many other ways you as a user can cause problems to your own
system...
--
Stephen J Smoogen.
"The core skill of innovators is error recovery, not failure avoidance."
Randy Nelson, President of Pixar University.
"Let us be kind, one to another, for most of us are fighting a hard
battle." -- Ian MacLaren
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, 10.05.11 01:04, Miloslav Trmač (m...@volny.cz) wrote: > > ... rereading the thread, the thing I have missed is that we don't > want to block startup in configurations where the server binds to > 127.0.0.1 because the system might never come online. So there appears > to be no universally correct default for programs with flexible bind > addresses :( >Mirek Note that the loopback device is actually configured as one of the first things in systemd, from within PID 1. On a systemd system 127.0.0.1 and ::1 are always bindable, the network scripts or NetworkManager do not need to set them up manually anymore. That basically means that on a systemd system a service can always bind to 0.0.0.0, ::, 127.0.0.1, ::1. Only if it needs a different IP address it can resort to IP_FREEBIND, or even better netlink notifications. Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, 10.05.11 00:57, Michał Piotrowski (mkkp...@gmail.com) wrote: > > 2011/5/10 Lennart Poettering : > > On Mon, 09.05.11 23:54, Michał Piotrowski (mkkp...@gmail.com) wrote: > >> No, only for /run/user/ - because there is a simple workaround that > >> can be used on affected systems if the administrator considers his > >> system as vulnerable for malicious users. > > > > Again, we had /dev/shm for years on Linux. This weakness in the security > > model is not news, not at all. > > Yes, but /run/user is a new thing and it gives wonderful opportunity > to DoS services for all system users. Thats my POV. And my POV is that > it should be documented - users should be aware about this. Also FPL > agreed with my arguments. > > That's all. If you do not agree with that, then I'm giving up :) All I am saying is that you can easily do this with /dev/shm too. And not only with F15, but with every prior release too, since /dev/shm was invented. Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Interlinux waiting list
Hello Thank you for your enquiry, Interlinux are currently operating a waiting list only facility for all services inquiries. Please download a free copy of our complementary ebooks: "Walking With The Elephants" http://diaser.org.uk/wwte/ebookwwte.zip about Open Source collaboration. and the public domain: "Damian's Social Networking Rule Book" http://diaser.org.uk/files/damiansnrules.pdf about the use of social networking media and tools. If you have an urgent Zentyal support request - please visit www.zentyla.com Both ebooks are available in Amazon Kindle format for £1.71 and £0.68. Kind Regards -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, 10.05.11 00:23, Miloslav Trmač (m...@volny.cz) wrote: > I can't see that. If I configure a server to listen on 192.168.1.1, > and my IP address changes to 192.168.1.2 for whatever reason, how will > I ever be notified that the configuration is incorrect? By a log entry > in syslog? If a local IP address becomes unavailable that a socket is bound to this has no effect on the socket itself, regardless whether IP_FREEBIND is used or not. i.e. if you managed to bind the socket once, it will stay bound for good; hence there is not really a difference in this regard. > * If an user logs in before the server starts, service status will be > reported as "pending" - which is correct > * After the network is up, the server can bind normally (without > relying on Linux-specific features). > * If the bind() fails, the server aborts and the service status will > be reported as "failed" - which is correct, and the obvious place to > check in case of problems. I think a robust system, regardless whether laptop or server should be as independent from the network as possible. That means it should provide its services under all circumstances, so that local clients can access it always, and remote services under all circumstances when the network is up. I think it is good design if your machine boots up no matter what, whether your ethernet cable is plugged or not, or whether your antenna broke, and it should always work as well as it can. If you hook into netlink changes then you can implement a fully robust server, and some servers work like this, for example bind. Using netlink is a very powerful tool but also not a one-line change in your server. A simpler solution is IP_FREEBIND or binding only to 0.0.0.0 (i.e. all local IP addresses). Or even combining 0.0.0.0 with SO_BINDTODEVICE to bind things by device instead of IP address. It's a simple fact that we live in a dynamic world these days. Hardware configuration changes, network configuration changes, all the time. We should write our software to deal with this as well as possible, and keep the dependencies we need to run at a minimum. That means that MySQL should not be delayed until the network is fully up, until a DHCP server replied. You want it run quickly, and then make the best of the network as it shows up, or as it goes away again, and even if it later returns. > Is the tradeoff really "correctness vs. saving a few seconds when > booting a server"? There is not really a tradeoff here. I think everybody wins if we make our systems robust, by not unnecessarily slowing down/allowing to hang our boot if external services like the network are not up. Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, May 10, 2011 at 12:52 AM, Jason D. Clinton wrote: > 2011/5/9 Miloslav Trmač : >> Is the tradeoff really "correctness vs. saving a few seconds when >> booting a server"? > > You know that's not what has been said and it's not really fair of > you. That was a question... > You want a server to fail on network failure; fine. Why don't you > just say that instead of characterizing it as "correctness"? I just can't see how a setup in which (systemctl status my_server) reports everything is fine and dandy when in fact the server is not able to accept a connection is correct, and especially "more robust" as it was claimed upthread. ... rereading the thread, the thing I have missed is that we don't want to block startup in configurations where the server binds to 127.0.0.1 because the system might never come online. So there appears to be no universally correct default for programs with flexible bind addresses :( Mirek -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/10 Lennart Poettering : > On Mon, 09.05.11 23:54, Michał Piotrowski (mkkp...@gmail.com) wrote: >> No, only for /run/user/ - because there is a simple workaround that >> can be used on affected systems if the administrator considers his >> system as vulnerable for malicious users. > > Again, we had /dev/shm for years on Linux. This weakness in the security > model is not news, not at all. Yes, but /run/user is a new thing and it gives wonderful opportunity to DoS services for all system users. Thats my POV. And my POV is that it should be documented - users should be aware about this. Also FPL agreed with my arguments. That's all. If you do not agree with that, then I'm giving up :) > > Lennart > > -- > Lennart Poettering - Red Hat, Inc. > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Miloslav Trmač : > Is the tradeoff really "correctness vs. saving a few seconds when > booting a server"? You know that's not what has been said and it's not really fair of you. You want a server to fail on network failure; fine. Why don't you just say that instead of characterizing it as "correctness"? -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Mon, 09.05.11 23:54, Michał Piotrowski (mkkp...@gmail.com) wrote: > > 2011/5/9 Lennart Poettering : > > On Mon, 09.05.11 22:46, Michał Piotrowski (mkkp...@gmail.com) wrote: > > > >> > >> 2011/5/9 Lennart Poettering : > >> > On Mon, 09.05.11 18:58, Michał Piotrowski (mkkp...@gmail.com) wrote: > >> > > >> >> >> Yes, mysql and postgresql are both certainly broken by this. Please > >> >> >> send the info, I'll take care of those two. > >> >> > > >> >> > I'll create bug reports and send proposed patches later. > >> >> > > >> >> > >> >> Patch for MySQL > >> >> https://bugzilla.redhat.com/show_bug.cgi?id=703214 > >> >> > >> >> for PostgreSQL > >> >> https://bugzilla.redhat.com/show_bug.cgi?id=703215 > >> >> > >> > > >> > Urks. I would strongly suggest not to make changes like this by > >> > default. These services should bind on 0.0.0.0 by default, which is > >> > available without network. > >> > >> Ok, when it comes to me - I can fix that on my setup :) > >> > >> But what about other large systems? > >> > >> When it comes to systemd in F15 a few things should be documented in > >> F15 release notes. /run vulnerability too. > > > > /run vulnerability? > > /run/user/ for accuracy > https://bugzilla.redhat.com/show_bug.cgi?id=693253 > > > > > Are you referring to the /dev/shm vulnerability? That's not really news... > > No, only for /run/user/ - because there is a simple workaround that > can be used on affected systems if the administrator considers his > system as vulnerable for malicious users. Again, we had /dev/shm for years on Linux. This weakness in the security model is not news, not at all. Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Tue, May 10, 2011 at 12:05 AM, Adam Williamson wrote: > You could have them wait for the network to be available simply by > setting them to depend on NetworkManager-wait-online.service: but that's > not a very good solution to anyone's problem. It's far better for them > to be able to start up before a network connection is available, and > then start working when one is available. It makes startup faster and > makes the servers in question more robust. I can't see that. If I configure a server to listen on 192.168.1.1, and my IP address changes to 192.168.1.2 for whatever reason, how will I ever be notified that the configuration is incorrect? By a log entry in syslog? It seems much more practical to me to set up such network-facing servers (which are not necessary for local user login) to only start after the network is available: * If an user logs in before the server starts, service status will be reported as "pending" - which is correct * After the network is up, the server can bind normally (without relying on Linux-specific features). * If the bind() fails, the server aborts and the service status will be reported as "failed" - which is correct, and the obvious place to check in case of problems. * On desktops nobody really cares how soon postgresql starts - being able to log quickly in is more important. * On servers with statically configured IP addresses, is the delay caused by NM-wait-online noticeable? * On systems that depend on DHCP, depending on NM-wait-online is necessary for reliable error detection. Is the tradeoff really "correctness vs. saving a few seconds when booting a server"? Mirek -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Mon, 2011-05-09 at 17:04 -0400, Tom Lane wrote: > > a) people who manually change the IP address fo mysql to bind on > > specific ip addresses, manually also enable > > NetworkManager-wait-online.service. > > > or: > > > b) the servers are fixed to listen to netlink. > > > or: > > > c) They get fixed to use IP_FREEBIND. > > > All three of the solutions are nicer than adding unnecessary > > dependencies for them. > > I think you've failed to grasp the point. What you are proposing is to I'd rather suggest you are. > hack the servers with patches that are rather unlikely to be accepted by > either upstream, Neither b) nor c) is a hack; they're both improvements in behaviour whether or not systemd is involved. It's simply more robust for a server to be able to run before the network connection is available (and hence across state changes). You don't explain why you think such a patch wouldn't be accepted upstream. > in order to solve *one* of the possible configuration > issues that might cause them to not start correctly before the basic > expected network support services are available. In particular, so far > as I can tell from the discussion at bug #703215, systemd is entirely > incapable of supporting services that need to do DNS lookups at start. That's not true; Tomasz Torcz wrote earlier in this thread "We have hackish NetworkManager-wait-online.service which can be requested in such cases." The point here is that the default, and probably most common, configuration of mysql and postgresql is for the server to bind to 0.0.0.0 (i.e. locally), so it's not the best thing for the initscript for these to wait for the network to be up, regardless of whether the service actually needs the network in its present configuration. You could have them wait for the network to be available simply by setting them to depend on NetworkManager-wait-online.service: but that's not a very good solution to anyone's problem. It's far better for them to be able to start up before a network connection is available, and then start working when one is available. It makes startup faster and makes the servers in question more robust. It's just a better solution all around. > I grow weary of systemd apologists saying that services should be hacked > to work around systemd's limitations. This is an incorrect interpretation of what's going on. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Lennart Poettering : > On Mon, 09.05.11 22:46, Michał Piotrowski (mkkp...@gmail.com) wrote: > >> >> 2011/5/9 Lennart Poettering : >> > On Mon, 09.05.11 18:58, Michał Piotrowski (mkkp...@gmail.com) wrote: >> > >> >> >> Yes, mysql and postgresql are both certainly broken by this. Please >> >> >> send the info, I'll take care of those two. >> >> > >> >> > I'll create bug reports and send proposed patches later. >> >> > >> >> >> >> Patch for MySQL >> >> https://bugzilla.redhat.com/show_bug.cgi?id=703214 >> >> >> >> for PostgreSQL >> >> https://bugzilla.redhat.com/show_bug.cgi?id=703215 >> >> >> > >> > Urks. I would strongly suggest not to make changes like this by >> > default. These services should bind on 0.0.0.0 by default, which is >> > available without network. >> >> Ok, when it comes to me - I can fix that on my setup :) >> >> But what about other large systems? >> >> When it comes to systemd in F15 a few things should be documented in >> F15 release notes. /run vulnerability too. > > /run vulnerability? /run/user/ for accuracy https://bugzilla.redhat.com/show_bug.cgi?id=693253 > > Are you referring to the /dev/shm vulnerability? That's not really news... No, only for /run/user/ - because there is a simple workaround that can be used on affected systems if the administrator considers his system as vulnerable for malicious users. > > Lennart > > -- > Lennart Poettering - Red Hat, Inc. > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
[389-devel] Please Review: (703304) Auto membership alternate config area should override default area
https://bugzilla.redhat.com/show_bug.cgi?id=703304 https://bugzilla.redhat.com/attachment.cgi?id=497928&action=edit -- 389-devel mailing list 389-de...@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/389-devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Mon, 09.05.11 17:04, Tom Lane (t...@redhat.com) wrote: > > Lennart Poettering writes: > > So, instead of making mysql/postgres start up slower for everybody, i'd > > rather see this solution: > > > a) people who manually change the IP address fo mysql to bind on > > specific ip addresses, manually also enable > > NetworkManager-wait-online.service. > > > or: > > > b) the servers are fixed to listen to netlink. > > > or: > > > c) They get fixed to use IP_FREEBIND. > > > All three of the solutions are nicer than adding unnecessary > > dependencies for them. > > I think you've failed to grasp the point. What you are proposing is to > hack the servers with patches that are rather unlikely to be accepted by > either upstream, in order to solve *one* of the possible configuration > issues that might cause them to not start correctly before the basic > expected network support services are available. In particular, so far > as I can tell from the discussion at bug #703215, systemd is entirely > incapable of supporting services that need to do DNS lookups at start. > It's not acceptable to tell people that they mustn't use configurations > that require that; they'll just go looking for another solution that > does do what they want. > > I grow weary of systemd apologists saying that services should be hacked > to work around systemd's limitations. systemd exists to serve the > daemons, not vice versa. If you can't fix these problems, people are > going to decide that systemd is a failed experiment. Hmm? which systemd problem in particular? I don't see how systemd changes anything in regards to networking here... Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Mon, 09.05.11 22:46, Michał Piotrowski (mkkp...@gmail.com) wrote: > > 2011/5/9 Lennart Poettering : > > On Mon, 09.05.11 18:58, Michał Piotrowski (mkkp...@gmail.com) wrote: > > > >> >> Yes, mysql and postgresql are both certainly broken by this. Please > >> >> send the info, I'll take care of those two. > >> > > >> > I'll create bug reports and send proposed patches later. > >> > > >> > >> Patch for MySQL > >> https://bugzilla.redhat.com/show_bug.cgi?id=703214 > >> > >> for PostgreSQL > >> https://bugzilla.redhat.com/show_bug.cgi?id=703215 > >> > > > > Urks. I would strongly suggest not to make changes like this by > > default. These services should bind on 0.0.0.0 by default, which is > > available without network. > > Ok, when it comes to me - I can fix that on my setup :) > > But what about other large systems? > > When it comes to systemd in F15 a few things should be documented in > F15 release notes. /run vulnerability too. /run vulnerability? Are you referring to the /dev/shm vulnerability? That's not really news... Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
Lennart Poettering writes: > So, instead of making mysql/postgres start up slower for everybody, i'd > rather see this solution: > a) people who manually change the IP address fo mysql to bind on > specific ip addresses, manually also enable > NetworkManager-wait-online.service. > or: > b) the servers are fixed to listen to netlink. > or: > c) They get fixed to use IP_FREEBIND. > All three of the solutions are nicer than adding unnecessary > dependencies for them. I think you've failed to grasp the point. What you are proposing is to hack the servers with patches that are rather unlikely to be accepted by either upstream, in order to solve *one* of the possible configuration issues that might cause them to not start correctly before the basic expected network support services are available. In particular, so far as I can tell from the discussion at bug #703215, systemd is entirely incapable of supporting services that need to do DNS lookups at start. It's not acceptable to tell people that they mustn't use configurations that require that; they'll just go looking for another solution that does do what they want. I grow weary of systemd apologists saying that services should be hacked to work around systemd's limitations. systemd exists to serve the daemons, not vice versa. If you can't fix these problems, people are going to decide that systemd is a failed experiment. regards, tom lane -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Lennart Poettering : > On Mon, 09.05.11 18:58, Michał Piotrowski (mkkp...@gmail.com) wrote: > >> >> Yes, mysql and postgresql are both certainly broken by this. Please >> >> send the info, I'll take care of those two. >> > >> > I'll create bug reports and send proposed patches later. >> > >> >> Patch for MySQL >> https://bugzilla.redhat.com/show_bug.cgi?id=703214 >> >> for PostgreSQL >> https://bugzilla.redhat.com/show_bug.cgi?id=703215 >> > > Urks. I would strongly suggest not to make changes like this by > default. These services should bind on 0.0.0.0 by default, which is > available without network. Ok, when it comes to me - I can fix that on my setup :) But what about other large systems? When it comes to systemd in F15 a few things should be documented in F15 release notes. /run vulnerability too. > Only if you manually configure them to bind > on a specific IP address you need the network up and configured. But not > even that is strictly necessary. A nice server uses netlink to subscribe > to network changes and binds as appropriate. A simpler server could even > invoke the IP_FREEBIND setsockopt (1-line patch), which allows them to > listen on the addresses before the interfaces are up (which you probably > want to do anyway, since it adds robustness and makes your server > independent from network changes). > > So, instead of making mysql/postgres start up slower for everybody, i'd > rather see this solution: > > a) people who manually change the IP address fo mysql to bind on > specific ip addresses, manually also enable > NetworkManager-wait-online.service. > > or: > > b) the servers are fixed to listen to netlink. > > or: > > c) They get fixed to use IP_FREEBIND. > > All three of the solutions are nicer than adding unnecessary > dependencies for them. > > Lennart > > -- > Lennart Poettering - Red Hat, Inc. > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Mon, 09.05.11 18:58, Michał Piotrowski (mkkp...@gmail.com) wrote: > >> Yes, mysql and postgresql are both certainly broken by this. Please > >> send the info, I'll take care of those two. > > > > I'll create bug reports and send proposed patches later. > > > > Patch for MySQL > https://bugzilla.redhat.com/show_bug.cgi?id=703214 > > for PostgreSQL > https://bugzilla.redhat.com/show_bug.cgi?id=703215 > Urks. I would strongly suggest not to make changes like this by default. These services should bind on 0.0.0.0 by default, which is available without network. Only if you manually configure them to bind on a specific IP address you need the network up and configured. But not even that is strictly necessary. A nice server uses netlink to subscribe to network changes and binds as appropriate. A simpler server could even invoke the IP_FREEBIND setsockopt (1-line patch), which allows them to listen on the addresses before the interfaces are up (which you probably want to do anyway, since it adds robustness and makes your server independent from network changes). So, instead of making mysql/postgres start up slower for everybody, i'd rather see this solution: a) people who manually change the IP address fo mysql to bind on specific ip addresses, manually also enable NetworkManager-wait-online.service. or: b) the servers are fixed to listen to netlink. or: c) They get fixed to use IP_FREEBIND. All three of the solutions are nicer than adding unnecessary dependencies for them. Lennart -- Lennart Poettering - Red Hat, Inc. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: karma for syslog-ng-3.2.3-2
On Sun, 2011-05-08 at 22:15 +0200, Matthias Runge wrote: > Greetings, > > we all know, the final change freeze arrives tomorrow, May 9th. > > I would like to ask you, if you could provide some (positive) karma for > syslog-ng. It is the latest stable version and I'd like to pull it into > F15 before change freeze. as it's not on the media, whether it makes final stable package set or not makes virtually no difference...just sayin :) -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: anaconda-15.31-1.fc15 - Update the requirements for memory..
On Sat, 2011-05-07 at 09:48 -0700, Kevin Higgins wrote: > >From what I can figure out it looks like lorax-0.4.4-1.fc15 uses and > needs more memory > I can run Fedora 15 in full shell on our slowest 1.5 Ghz laptop with > 256mb memory and it runs faster > than f14, I do an install with 512MB than take out stick. The live cd > as of May6 that I made using > the repo.../development/15/i386/os/ only, does a full install without > problems, it doesn't seem to use > lorax for the install. But if we make the minimum 768 for anaconda > than I will no longer be able > to even do a livecd install to these machines, that we have since > 2004, installed > Fedora on them and than given them away to needy families in our area. > Is there a way to have anaconda have a minimum requirements for > install and a separate minimum > requirements for a live install? These systems are donated to us we > refurbish them and then they go back out > loaded with Fedora to students and families in need. this is already known and discussed at: https://bugzilla.redhat.com/show_bug.cgi?id=680542 https://bugzilla.redhat.com/show_bug.cgi?id=682555 https://bugzilla.redhat.com/show_bug.cgi?id=696805 https://bugzilla.redhat.com/show_bug.cgi?id=499585 -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Interlinux waiting list
Hello Thank you for your enquiry, Interlinux are currently operating a waiting list only facility for all services inquiries. Please download a free copy of our complementary ebooks: "Walking With The Elephants" http://diaser.org.uk/wwte/ebookwwte.zip about Open Source collaboration. and the public domain: "Damian's Social Networking Rule Book" http://diaser.org.uk/files/damiansnrules.pdf about the use of social networking media and tools. If you have an urgent Zentyal support request - please visit www.zentyla.com Both ebooks are available in Amazon Kindle format for £1.71 and £0.68. Kind Regards -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
=?ISO-8859-2?Q?Micha=B3_Piotrowski?= writes: > 2011/5/9 Tom Lane : >> I'd be interested to know whether the OP was using a vanilla my.cnf or >> something custom, and exactly what failure mode he saw. > In my case > [mysqld] > datadir=/home/data/mysql > socket=/var/lib/mysql/mysql.sock > user=mysql > # Disabling symbolic-links is recommended to prevent assorted security risks > symbolic-links=0 > bind-address = 192.168.101.200 Ah, I'll bet this bind-address setting is what's doing it --- if that address weren't configured yet, it'd be unsurprising that mysqld would fail to start. (The comparable case I tested in postgresql was to set listen_addresses = 'full.name.of.host', which failed due to DNS not working yet. But now that I look at this example, I bet setting listen_addresses to a numeric IP would have the same issue as your mysql example.) Thanks for the followup. regards, tom lane -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Tom Lane : > Reindl Harald writes: >> Am 09.05.2011 19:09, schrieb Michał Piotrowski: >>> W dniu 9 maja 2011 18:10 użytkownik Michal Schmidt >>> napisał: Are you sure that none of these services can deal with the network connection coming up later? > >>> I don't know. It appears that MySQL can't deal with that > >> i wonder that mysqld should need network in every case since most setups >> out there using >> only sockets for mysql-connections and it is a valid use-case having mysql >> on machines >> without any network > > I did some quick tests and it seemed that the plain-vanilla > configuration of both postgres and mysql would start all right in F15. > However, it's definitely possible to configure postgres in a way that > requires it to do DNS lookups at startup, and I verified that that will > fail in the current state of F15. I'm less familiar with all of mysql's > options but I wouldn't be surprised if there's comparable issues there. > > I'd be interested to know whether the OP was using a vanilla my.cnf or > something custom, and exactly what failure mode he saw. In my case [mysqld] datadir=/home/data/mysql socket=/var/lib/mysql/mysql.sock user=mysql # Disabling symbolic-links is recommended to prevent assorted security risks symbolic-links=0 bind-address = 192.168.101.200 init_connect='SET collation_connection = utf8_general_ci' init_connect='SET NAMES utf8' character-set-server = utf8 collation-server = utf8_general_ci [mysql] default-character-set=utf8 [mysqld_safe] log-error=/var/log/mysqld.log pid-file=/var/run/mysqld/mysqld.pid MySQL exited with code 1 mysqld.service - SYSV: MySQL database server. Loaded: loaded (/etc/rc.d/init.d/mysqld) Active: failed since Mon, 09 May 2011 08:15:44 +0200; 4min 32s ago Process: 549 ExecStart=/etc/rc.d/init.d/mysqld start (code=exited, status=1/FAILURE) CGroup: name=systemd:/system/mysqld.service Here is a fragment from log 110509 8:15:41 [Note] Plugin 'FEDERATED' is disabled. 110509 8:15:41 InnoDB: The InnoDB memory heap is disabled 110509 8:15:41 InnoDB: Mutexes and rw_locks use GCC atomic builtins 110509 8:15:41 InnoDB: Compressed tables use zlib 1.2.5 110509 8:15:41 InnoDB: Using Linux native AIO 110509 8:15:41 InnoDB: Initializing buffer pool, size = 128.0M 110509 8:15:41 InnoDB: Completed initialization of buffer pool 110509 8:15:41 InnoDB: highest supported file format is Barracuda. 110509 8:15:42 InnoDB: Waiting for the background threads to start 110509 8:15:43 InnoDB: 1.1.5 started; log sequence number 48965389142 110509 8:15:43 [ERROR] Can't start server: Bind on TCP/IP port: Cannot assign requested address 110509 8:15:43 [ERROR] Do you already have another mysqld server running on port: 3306 ? 110509 8:15:43 [ERROR] Aborting 110509 8:15:43 InnoDB: Starting shutdown... 110509 8:15:44 InnoDB: Shutdown completed; log sequence number 48965389142 110509 8:15:44 [Note] /usr/libexec/mysqld: Shutdown complete > > regards, tom lane > > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel > -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
Reindl Harald writes: > Am 09.05.2011 19:09, schrieb Micha³ Piotrowski: >> W dniu 9 maja 2011 18:10 u¿ytkownik Michal Schmidt >> napisa³: >>> Are you sure that none of these services can deal with the network >>> connection coming up later? >> I don't know. It appears that MySQL can't deal with that > i wonder that mysqld should need network in every case since most setups out > there using > only sockets for mysql-connections and it is a valid use-case having mysql > on machines > without any network I did some quick tests and it seemed that the plain-vanilla configuration of both postgres and mysql would start all right in F15. However, it's definitely possible to configure postgres in a way that requires it to do DNS lookups at startup, and I verified that that will fail in the current state of F15. I'm less familiar with all of mysql's options but I wouldn't be surprised if there's comparable issues there. I'd be interested to know whether the OP was using a vanilla my.cnf or something custom, and exactly what failure mode he saw. regards, tom lane -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
W dniu 9 maja 2011 19:36 użytkownik Reindl Harald napisał: > i wonder that mysqld should need network in every case since most setups out > there using > only sockets for mysql-connections and it is a valid use-case having mysql > on machines > without any network Good question - indeed, so it should be. -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
Am 09.05.2011 19:09, schrieb Michał Piotrowski: > W dniu 9 maja 2011 18:10 użytkownik Michal Schmidt > napisał: >> On 05/09/2011 05:02 PM, Michał Piotrowski wrote: >>> Ok, I know what is happening. I looked at the init script and I >>> realized that it doesn't have any LSB header info about boot process >>> order. I also looked at other services and here are some of them that >>> needs fixing: >>> - memcached - (I created systemd service for it) - I can provide a >>> patch for classic init script >>> - smb - as above >>> - nmb - as above >>> - mysql - I can provide a patch for classic init scrip >>> - postgresql - as above >> >> Are you sure that none of these services can deal with the network >> connection coming up later? > > I don't know. It appears that MySQL can't deal with that i wonder that mysqld should need network in every case since most setups out there using only sockets for mysql-connections and it is a valid use-case having mysql on machines without any network signature.asc Description: OpenPGP digital signature -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
W dniu 9 maja 2011 19:21 użytkownik Orion Poplawski napisał: > On 05/09/2011 10:10 AM, Michal Schmidt wrote: >> On 05/09/2011 05:02 PM, Michał Piotrowski wrote: >>> Ok, I know what is happening. I looked at the init script and I >>> realized that it doesn't have any LSB header info about boot process >>> order. I also looked at other services and here are some of them that >>> needs fixing: >>> - memcached - (I created systemd service for it) - I can provide a >>> patch for classic init script >>> - smb - as above >>> - nmb - as above >>> - mysql - I can provide a patch for classic init scrip >>> - postgresql - as above >> >> Are you sure that none of these services can deal with the network >> connection coming up later? >> >> Michal > > nmb can't: https://bugzilla.redhat.com/show_bug.cgi?id=486231 Here you can find the fix :) https://bugzilla.redhat.com/show_bug.cgi?id=660069 > > -- > Orion Poplawski > Technical Manager 303-415-9701 x222 > NWRA/CoRA Division FAX: 303-415-9702 > 3380 Mitchell Lane or...@cora.nwra.com > Boulder, CO 80301 http://www.cora.nwra.com > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel > -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On 05/09/2011 10:10 AM, Michal Schmidt wrote: > On 05/09/2011 05:02 PM, Michał Piotrowski wrote: >> Ok, I know what is happening. I looked at the init script and I >> realized that it doesn't have any LSB header info about boot process >> order. I also looked at other services and here are some of them that >> needs fixing: >> - memcached - (I created systemd service for it) - I can provide a >> patch for classic init script >> - smb - as above >> - nmb - as above >> - mysql - I can provide a patch for classic init scrip >> - postgresql - as above > > Are you sure that none of these services can deal with the network > connection coming up later? > > Michal nmb can't: https://bugzilla.redhat.com/show_bug.cgi?id=486231 -- Orion Poplawski Technical Manager 303-415-9701 x222 NWRA/CoRA DivisionFAX: 303-415-9702 3380 Mitchell Lane or...@cora.nwra.com Boulder, CO 80301 http://www.cora.nwra.com -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
W dniu 9 maja 2011 18:10 użytkownik Michal Schmidt napisał: > On 05/09/2011 05:02 PM, Michał Piotrowski wrote: >> Ok, I know what is happening. I looked at the init script and I >> realized that it doesn't have any LSB header info about boot process >> order. I also looked at other services and here are some of them that >> needs fixing: >> - memcached - (I created systemd service for it) - I can provide a >> patch for classic init script >> - smb - as above >> - nmb - as above >> - mysql - I can provide a patch for classic init scrip >> - postgresql - as above > > Are you sure that none of these services can deal with the network > connection coming up later? I don't know. It appears that MySQL can't deal with that. > > Michal > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel > -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Tomasz Torcz : > On Mon, May 09, 2011 at 11:26:50AM -0400, Tom Lane wrote: >> =?ISO-8859-2?Q?Micha=B3_Piotrowski?= writes: >> > Ok, I know what is happening. I looked at the init script and I >> > realized that it doesn't have any LSB header info about boot process >> > order. >> >> No, it's assuming that the chkconfig numbering takes care of that. >> Are you saying that systemd no longer honors the chkconfig ordering? >> That doesn't leave me with a warm feeling about how much testing the >> systemd stuff has gotten, because there is all kinds of stuff that's >> liable to fall over without that. > > Systemd ignores chkconfig ordering for native services. And orders > by requirements. So it is crucial to have proper Requires= and After= > in unit files. Unfortunately I did not have time to create native services for PgSQL and MySQL, so I need to use old sysvscripts - therefore, I noticed this problem. OTOH these are the last two sysvinit scripts that I use - so I probably did not noticed more problems with other things :) I wonder how many of these problems arises in large systems after upgrade to F15 > >> > I also looked at other services and here are some of them that >> > needs fixing: >> > - memcached - (I created systemd service for it) - I can provide a >> > patch for classic init script >> > - smb - as above >> > - nmb - as above >> > - mysql - I can provide a patch for classic init scrip >> > - postgresql - as above >> >> Yes, mysql and postgresql are both certainly broken by this. Please >> send the info, I'll take care of those two. > > We have hackish NetworkManager-wait-online.service which can be requested > in such cases. But it really shouldn't be used. Services should have > proper requirements set, deal with dynamic network nature. And ideally > by socket activable, so we don't have to deal with such issues. > > -- > Tomasz Torcz RIP is irrevelant. Spoofing is futile. > xmpp: zdzich...@chrome.pl Your routes will be aggreggated. -- Alex Yuriev > > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel > -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
[perl-Devel-PatchPerl/el6/master] excludearch ppc64 because it's missing perl(File::pushd)
commit 3279e4952de0582d2c35b9eba7a3cf909f207043
Author: Iain Arnell
Date: Mon May 9 18:19:12 2011 +0200
excludearch ppc64 because it's missing perl(File::pushd)
perl-Devel-PatchPerl.spec |7 ++-
1 files changed, 6 insertions(+), 1 deletions(-)
---
diff --git a/perl-Devel-PatchPerl.spec b/perl-Devel-PatchPerl.spec
index ce76381..e40ca79 100644
--- a/perl-Devel-PatchPerl.spec
+++ b/perl-Devel-PatchPerl.spec
@@ -1,6 +1,6 @@
Name: perl-Devel-PatchPerl
Version:0.30
-Release:1%{?dist}
+Release:1%{?dist}.1
Summary:Patch perl source à la Devel::PPPort's buildperl.pl
License:GPL+ or Artistic
Group: Development/Libraries
@@ -17,6 +17,8 @@ BuildRequires: perl(Test::More)
Requires: perl(File::pushd) >= 1.00
Requires: perl(IPC::Cmd) >= 0.40
Requires: perl(:MODULE_COMPAT_%(eval "`%{__perl} -V:version`"; echo
$version))
+# perl(File::pushd) isn't available in RHEL6 ppc64
+ExcludeArch:ppc64
%{?perl_default_filter}
@@ -51,5 +53,8 @@ make test
%{_mandir}/man3/*
%changelog
+* Mon May 09 2011 Iain Arnell 0.30-1.1
+- excludearch ppc64 because it's missing perl(File::pushd)
+
* Wed Apr 27 2011 Iain Arnell 0.30-1
- Specfile autogenerated by cpanspec 1.78.
--
Fedora Extras Perl SIG
http://www.fedoraproject.org/wiki/Extras/SIGs/Perl
perl-devel mailing list
perl-de...@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/perl-devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On 05/09/2011 05:02 PM, Michał Piotrowski wrote: > Ok, I know what is happening. I looked at the init script and I > realized that it doesn't have any LSB header info about boot process > order. I also looked at other services and here are some of them that > needs fixing: > - memcached - (I created systemd service for it) - I can provide a > patch for classic init script > - smb - as above > - nmb - as above > - mysql - I can provide a patch for classic init scrip > - postgresql - as above Are you sure that none of these services can deal with the network connection coming up later? Michal -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
W dniu 9 maja 2011 17:34 użytkownik Michał Piotrowski napisał: > 2011/5/9 Tom Lane : >> =?ISO-8859-2?Q?Micha=B3_Piotrowski?= writes: >>> Ok, I know what is happening. I looked at the init script and I >>> realized that it doesn't have any LSB header info about boot process >>> order. >> >> No, it's assuming that the chkconfig numbering takes care of that. > > Do you mean these numbers before script name in /etc/rc.d/rc*.d/ ? > >> Are you saying that systemd no longer honors the chkconfig ordering? > > It seems to me that mysql starts before NetworkManager here > http://eventhorizon.pl/images/bootchart17.png > >> That doesn't leave me with a warm feeling about how much testing the >> systemd stuff has gotten, because there is all kinds of stuff that's >> liable to fall over without that. >> >>> I also looked at other services and here are some of them that >>> needs fixing: >>> - memcached - (I created systemd service for it) - I can provide a >>> patch for classic init script >>> - smb - as above >>> - nmb - as above >>> - mysql - I can provide a patch for classic init scrip >>> - postgresql - as above >> >> Yes, mysql and postgresql are both certainly broken by this. Please >> send the info, I'll take care of those two. > > I'll create bug reports and send proposed patches later. > Patch for MySQL https://bugzilla.redhat.com/show_bug.cgi?id=703214 for PostgreSQL https://bugzilla.redhat.com/show_bug.cgi?id=703215 Bootchart shows me that now MySQL starts in the right order. What about the rest of the mentioned init scripts? For the samba there is a native systemd service in bugzilla - the same for memcached. -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
On Mon, May 09, 2011 at 11:26:50AM -0400, Tom Lane wrote: > =?ISO-8859-2?Q?Micha=B3_Piotrowski?= writes: > > Ok, I know what is happening. I looked at the init script and I > > realized that it doesn't have any LSB header info about boot process > > order. > > No, it's assuming that the chkconfig numbering takes care of that. > Are you saying that systemd no longer honors the chkconfig ordering? > That doesn't leave me with a warm feeling about how much testing the > systemd stuff has gotten, because there is all kinds of stuff that's > liable to fall over without that. Systemd ignores chkconfig ordering for native services. And orders by requirements. So it is crucial to have proper Requires= and After= in unit files. > > I also looked at other services and here are some of them that > > needs fixing: > > - memcached - (I created systemd service for it) - I can provide a > > patch for classic init script > > - smb - as above > > - nmb - as above > > - mysql - I can provide a patch for classic init scrip > > - postgresql - as above > > Yes, mysql and postgresql are both certainly broken by this. Please > send the info, I'll take care of those two. We have hackish NetworkManager-wait-online.service which can be requested in such cases. But it really shouldn't be used. Services should have proper requirements set, deal with dynamic network nature. And ideally by socket activable, so we don't have to deal with such issues. -- Tomasz Torcz RIP is irrevelant. Spoofing is futile. xmpp: zdzich...@chrome.pl Your routes will be aggreggated. -- Alex Yuriev -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
2011/5/9 Tom Lane : > =?ISO-8859-2?Q?Micha=B3_Piotrowski?= writes: >> Ok, I know what is happening. I looked at the init script and I >> realized that it doesn't have any LSB header info about boot process >> order. > > No, it's assuming that the chkconfig numbering takes care of that. Do you mean these numbers before script name in /etc/rc.d/rc*.d/ ? > Are you saying that systemd no longer honors the chkconfig ordering? It seems to me that mysql starts before NetworkManager here http://eventhorizon.pl/images/bootchart17.png > That doesn't leave me with a warm feeling about how much testing the > systemd stuff has gotten, because there is all kinds of stuff that's > liable to fall over without that. > >> I also looked at other services and here are some of them that >> needs fixing: >> - memcached - (I created systemd service for it) - I can provide a >> patch for classic init script >> - smb - as above >> - nmb - as above >> - mysql - I can provide a patch for classic init scrip >> - postgresql - as above > > Yes, mysql and postgresql are both certainly broken by this. Please > send the info, I'll take care of those two. I'll create bug reports and send proposed patches later. > > regards, tom lane > -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: informations about boot sequence (Re: F15 - mysql start problem)
=?ISO-8859-2?Q?Micha=B3_Piotrowski?= writes: > Ok, I know what is happening. I looked at the init script and I > realized that it doesn't have any LSB header info about boot process > order. No, it's assuming that the chkconfig numbering takes care of that. Are you saying that systemd no longer honors the chkconfig ordering? That doesn't leave me with a warm feeling about how much testing the systemd stuff has gotten, because there is all kinds of stuff that's liable to fall over without that. > I also looked at other services and here are some of them that > needs fixing: > - memcached - (I created systemd service for it) - I can provide a > patch for classic init script > - smb - as above > - nmb - as above > - mysql - I can provide a patch for classic init scrip > - postgresql - as above Yes, mysql and postgresql are both certainly broken by this. Please send the info, I'll take care of those two. regards, tom lane -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
informations about boot sequence (Re: F15 - mysql start problem)
W dniu 9 maja 2011 09:54 użytkownik Michał Piotrowski napisał: > W dniu 9 maja 2011 09:51 użytkownik Michał Piotrowski > napisał: >> 2011/5/9 Nicolas Mailhot : >>> >>> >>> Le Lun 9 mai 2011 08:32, Michał Piotrowski a écrit : >>> It seems to me that this is related to that at the time of starting the network is not available. IIRC there was a problem with NetworkManager and it's systemd service, but it was fixed AFAIK. There is no problem with other services. Any ideas what could go wrong? >>> >>> Most services that depend on Network want all the networking bits available >>> (dns, proxy if the system is using a proxy, etc). >> >> mysqld starts earlier than NetworkManager here >> http://i51.tinypic.com/66gn61.png > > Here is a better quality > http://eventhorizon.pl/images/bootchart17.png > Ok, I know what is happening. I looked at the init script and I realized that it doesn't have any LSB header info about boot process order. I also looked at other services and here are some of them that needs fixing: - memcached - (I created systemd service for it) - I can provide a patch for classic init script - smb - as above - nmb - as above - mysql - I can provide a patch for classic init scrip - postgresql - as above -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: Is Rawhide supposed to be useful?
On 09/05/11 13:49, Matthew Miller wrote: To answer the subject, for testing it is usefull. For deploying satellites, not so much. I have two rawhide vms' running But I find with rawhide. Keep rpm copies local with yum*local. for rawhide I keep 12 copes of all installed rpms. yum downgrade xorg*server* I have x all the time as a result. -- Regards, Frank Murphy UTF_8 Encoded Friend of Fedora -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Is Rawhide supposed to be useful?
https://bugzilla.redhat.com/show_bug.cgi?id=698237 and it's been this way for a while: http://article.gmane.org/gmane.linux.redhat.fedora.devel/147699 which just says Rawhide's X is going to be like that for a while. We have ABI version checks in place, but they only operate off the numbers actually exported from the server, and those don't change until ABI freeze for the server release. - ajax which is not very helpful -- and that was several weeks ago. The bug above keeps accumlating duplicates. Can some resources be put into making sure this is resolved and protected against via RPM dependencies? -- Matthew Miller Senior Systems Architect -- Instructional & Research Computing Services Harvard School of Engineering & Applied Sciences -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
rawhide report: 20110509 changes
Compose started at Mon May 9 08:15:02 UTC 2011 Broken deps for x86_64 -- beldi-0.9.25-3.fc15.x86_64 requires libhal-storage.so.1()(64bit) beldi-0.9.25-3.fc15.x86_64 requires libhal.so.1()(64bit) beldi-0.9.25-3.fc15.x86_64 requires hal callweaver-javascript-1.2.1-8.fc16.x86_64 requires libjs.so.1()(64bit) camcardsync-0.1.1-4.fc15.x86_64 requires libhal.so.1()(64bit) couchdb-1.0.2-2.fc16.x86_64 requires libjs.so.1()(64bit) cyphesis-0.5.24-3.fc15.x86_64 requires libskstream-0.3.so.4()(64bit) db4o-7.4-2.fc13.x86_64 requires mono(Mono.GetOptions) = 0:2.0.0.0 dh-make-0.55-3.fc15.noarch requires debhelper ember-0.6.0-6.fc16.x86_64 requires libskstream-0.3.so.4()(64bit) eris-1.3.16-2.fc15.i686 requires libskstream-0.3.so.4 eris-1.3.16-2.fc15.x86_64 requires libskstream-0.3.so.4()(64bit) exaile-0.3.2.1-1.fc16.noarch requires hal fence-virtd-libvirt-qpid-0.2.1-4.fc15.x86_64 requires libqmfconsole.so.3()(64bit) file-browser-applet-0.6.6-1.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) gedit-valencia-0.3.0-4.fc14.x86_64 requires libvala-0.10.so.0()(64bit) glunarclock-0.34.1-1.fc14.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-bubblemon-2.0.15-1.fc13.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-cpufire-1.6-3.fc14.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-globalmenu-0.7.9-1.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-grandr-0.4.1-2.fc12.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-music-2.5.1-5.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-sensors-2.2.7-4.fc15.i686 requires libpanel-applet-2.so.0 gnome-applet-sensors-2.2.7-4.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-applet-sshmenu-3.18-3.fc15.noarch requires ruby(panelapplet2) gnome-applet-window-picker-0.5.8-2.fc14.x86_64 requires libpanel-applet-2.so.0()(64bit) 1:gnome-applets-2.32.0-3.fc15.x86_64 requires libpanel-applet-3.so.0()(64bit) 1:gnome-applets-2.32.0-3.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) 1:gnome-applets-2.32.0-3.fc15.x86_64 requires libgweather.so.1()(64bit) gnome-device-manager-0.2-6.fc15.x86_64 requires libhal.so.1()(64bit) gnome-device-manager-devel-0.2-6.fc15.i686 requires hal-devel >= 0:0.5.10 gnome-device-manager-devel-0.2-6.fc15.i686 requires pkgconfig(hal) gnome-device-manager-devel-0.2-6.fc15.x86_64 requires hal-devel >= 0:0.5.10 gnome-device-manager-devel-0.2-6.fc15.x86_64 requires pkgconfig(hal) gnome-device-manager-libs-0.2-6.fc15.i686 requires libhal.so.1 gnome-device-manager-libs-0.2-6.fc15.i686 requires hal >= 0:0.5.10 gnome-device-manager-libs-0.2-6.fc15.x86_64 requires libhal.so.1()(64bit) gnome-device-manager-libs-0.2-6.fc15.x86_64 requires hal >= 0:0.5.10 gnome-netstatus-2.28.2-1.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-pilot-conduits-2.32.1-2.fc15.x86_64 requires libgpilotdconduit.so.3()(64bit) gnome-pilot-conduits-2.32.1-2.fc15.x86_64 requires libgpilotd.so.5()(64bit) gnome-pilot-conduits-2.32.1-2.fc15.x86_64 requires libgpilotdcm.so.4()(64bit) gnome-python2-applet-2.32.0-1.fc14.x86_64 requires libpanel-applet-2.so.0()(64bit) gnome-python2-brasero-2.32.0-1.fc14.x86_64 requires libbrasero-burn.so.1()(64bit) gnome-python2-brasero-2.32.0-1.fc14.x86_64 requires libbrasero-media.so.1()(64bit) gnome-python2-evince-2.32.0-1.fc14.x86_64 requires libevview.so.3()(64bit) gnome-python2-evince-2.32.0-1.fc14.x86_64 requires libevdocument.so.3()(64bit) gnome-python2-evolution-2.32.0-1.fc14.x86_64 requires libcamel-1.2.so.19()(64bit) gnome-python2-gdl-2.25.3-22.fc15.x86_64 requires libgdl-1.so.3()(64bit) gnome-python2-totem-2.32.0-1.fc14.x86_64 requires libgnome-media-profiles.so.0()(64bit) gnome-rdp-0.2.3-6.fc12.x86_64 requires mono(Mono.Data.SqliteClient) = 0:2.0.0.0 gnotime-2.3.0-8.fc15.x86_64 requires libgtkhtml-3.15.so.19()(64bit) gnubiff-2.2.13-4.fc15.x86_64 requires libpanel-applet-2.so.0()(64bit) gold-2.1.12.2-5.fc15.noarch requires perl(Data::Properties) gorm-1.2.13-0.20110331.fc16.i686 requires libgnustep-gui.so.0.19 gorm-1.2.13-0.20110331.fc16.i686 requires libgnustep-base.so.1.21 gorm-1.2.13-0.20110331.fc16.x86_64 requires libgnustep-gui.so.0.19()(64bit) gorm-1.2.13-0.20110331.fc16.x86_64 requires libgnustep-base.so.1.21()(64bit) gpx-viewer-0.2.0-3.fc14.x86_64 requires libchamplain-0.6.so.0()(64bit) gpx-viewer-0.2.0-3.fc14.x86_64 requires libchamplain-gtk-0.6.so.0()(64bit) gpx-viewer-0.2.0-3.fc14.x86_64 requires
Re: F15 - mysql start problem
W dniu 9 maja 2011 09:51 użytkownik Michał Piotrowski napisał: > 2011/5/9 Nicolas Mailhot : >> >> >> Le Lun 9 mai 2011 08:32, Michał Piotrowski a écrit : >> >>> It seems to me that this is related to that at the time of starting >>> the network is not available. IIRC there was a problem with >>> NetworkManager and it's systemd service, but it was fixed AFAIK. There >>> is no problem with other services. Any ideas what could go wrong? >> >> Most services that depend on Network want all the networking bits available >> (dns, proxy if the system is using a proxy, etc). > > mysqld starts earlier than NetworkManager here > http://i51.tinypic.com/66gn61.png Here is a better quality http://eventhorizon.pl/images/bootchart17.png > >> However systemd seems to >> treat "network available" as "link is up". Other stuff will come up, try to >> resolve something (typically, hostname for logs), and fail >> >> At least, that's one common error case I know of >> >> -- >> Nicolas Mailhot >> >> >> -- >> devel mailing list >> devel@lists.fedoraproject.org >> https://admin.fedoraproject.org/mailman/listinfo/devel > > > > -- > Best regards, > Michal > > http://eventhorizon.pl/ > -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: F15 - mysql start problem
2011/5/9 Nicolas Mailhot : > > > Le Lun 9 mai 2011 08:32, Michał Piotrowski a écrit : > >> It seems to me that this is related to that at the time of starting >> the network is not available. IIRC there was a problem with >> NetworkManager and it's systemd service, but it was fixed AFAIK. There >> is no problem with other services. Any ideas what could go wrong? > > Most services that depend on Network want all the networking bits available > (dns, proxy if the system is using a proxy, etc). mysqld starts earlier than NetworkManager here http://i51.tinypic.com/66gn61.png > However systemd seems to > treat "network available" as "link is up". Other stuff will come up, try to > resolve something (typically, hostname for logs), and fail > > At least, that's one common error case I know of > > -- > Nicolas Mailhot > > > -- > devel mailing list > devel@lists.fedoraproject.org > https://admin.fedoraproject.org/mailman/listinfo/devel -- Best regards, Michal http://eventhorizon.pl/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Re: F15 - mysql start problem
Le Lun 9 mai 2011 08:32, Michał Piotrowski a écrit : > It seems to me that this is related to that at the time of starting > the network is not available. IIRC there was a problem with > NetworkManager and it's systemd service, but it was fixed AFAIK. There > is no problem with other services. Any ideas what could go wrong? Most services that depend on Network want all the networking bits available (dns, proxy if the system is using a proxy, etc). However systemd seems to treat "network available" as "link is up". Other stuff will come up, try to resolve something (typically, hostname for logs), and fail At least, that's one common error case I know of -- Nicolas Mailhot -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
