Re: Honest Questions -- Trying to decide on version of linux/os
On 5/19/21 6:06 PM, bruce wrote: Hi. In the middle of trying to figure out a dev platform. So, looking at a blank slate to figure out what version of OS should have on the "work" laptop. Project work will be on the laptop as well as cloud VM.. In general, until not too long ago, Fedora used to be an excellent choice for dev-purposes. Unfortunately, this does not apply anymore, due to some strategic decisions Fedora and RHAT have taken, IMO. At the same time (if it matters), looking to to have a new laptop -- 12-16G, 256GSSD/1TB, 6-8 core. I'd prefer to have an OS that's rather stable, as I'm not looking to be the "sys admin" role.. Although, if I can find a remote sys admin that I could trust for a reasonable fee. I could be down for it. Given that this is Fed, I expect the maority of replies will lean to Fed! So, I'm interested in thoughts for the group. As much as I hate to say so, as a long term Fedora user, ATM, probably some Ubuntu variant is the best choice. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
Re: Strange error with 'man'
On 12/7/20 2:15 PM, Patrick O'Callaghan wrote: $ rpm -qd lutris /usr/share/man/lutris.1 [poc@Bree ~]$ man lutris No manual entry for lutris [poc@Bree ~]$ ls -l /usr/share/man/lutris.1 -rw-r--r--. 1 root root 1433 Jul 19 00:02 /usr/share/man/lutris.1 I also ran 'mandb' just in case. No difference. No surprise. This man page is installed into a bogus directory, i.e. this is a packaging bug. File a bug against this package to have it fixed. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: mysterious/suspicious internet activity.
On 12/1/20 1:18 AM, home user wrote: (on Mon, 2020-11-30 at 23:56 +, Ed wrote) > I thought you said your system was "quiet"? > > For your "network activity" issue the lines of interest are those > which include "ESTABLISHED" as the state. > > It shows both "thunderbird" and "firefox" are both running and connected > to hosts. So, one would expect some network activity When I opened the thread this morning (hours ago), my system was quiet. When, several minutes ago, I did the netstat -atuevp whose output I attached to my reply to you and Tim, yes, Thunderbird and Firefox were running. Check these packages' "Preferences->Privacy & Security" settings. IIRC, Fedora's Mozilla packages have Mozilla's espionage (IMHO: mis-) features (aka. "telemetry", "data collection") enabled by default and therefore are phoning home under "the hood". Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Policy regarding opt-out telemetry and privacy
On 10/27/20 9:57 PM, Olivier Lemasle wrote: Hi all, I'm packaging Open Policy Agent [1] (OPA) for Fedora. However, with version 0.20.0, OPA added a telemetry service, enabled by default, reporting to a OPA-managed service the OPA version, a UUID and the build architecture (cf changelog [2] and privacy information [3]) I didn't find any Fedora policy regarding this kind of opt-out telemetry, so I asked the Fedora Packaging Commitee for advice [4]. I got advised to ask Fedora community on this mailing list. So do you think it is ok to package OPA as is, or should I patch it to make telemetry opt-in by disabling it by default in the Fedora package? More globally, what do you think should be done in Fedora packages when an upstream project includes a telemetry service? Fedora has always obeyed a "no phone home" policy, i.e. "no telemetry" or other means of espionage by default. Besides this, any "by default active telemetry" would likely be unlawful in the EU, because it violates the GDPR[1] Ralf [1] https://en.wikipedia.org/wiki/General_Data_Protection_Regulation ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Fedora 30 EOL
On 5/28/20 1:30 PM, Patrick O'Callaghan wrote: On Thu, 2020-05-28 at 18:08 +0930, Tim via users wrote: Perhaps there should be an automated culling of participants. If you step up the plate to say you'll maintain a package, but don't, *you* get dumped from bugzilla. I was about to suggest something similar. Not necessarily dumping them from BZ but removing the package's "maintained" status (or whatever the terminology is). Clearly if bug reports are not being addressed, the package is not being maintained in any meaningful sense. Please do yourself a favor and check which packages and which packages you talking about. As I wrote before, there are obvious patterns, but we all know the people in charge are not interested. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Fedora 30 EOL
On 5/27/20 9:42 PM, Ben Cotton wrote: On Wed, May 27, 2020 at 3:23 PM Michael Schwendt wrote: The fundamental problem here is that it has taken a very long time for somebody to respond to the bug reporter. There has been no guidance and no hint whether anyone "somewhere" would be interested in looking into this issue. Exactly. And growing the contributor community is how we can solve that problem. Is this more than a cheap excuse? This problem exists ever since Fedora exists and has never been addressed. I agree with Michael. The fundamental problem is many maintainers not responding in timely manners (when a problem still is acute to the reporter), but maintainers trying to "sit out" problems until garbage collection sweeps them under the carpet. Furthermore, when you have a closer look at which BZs don't get addressed you'll find pretty obvious patterns, for the reasons why they have not been addressed. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: today's f32 update hung on restorecon
On 5/6/20 3:37 PM, Ed Greshko wrote: On 2020-05-06 20:30, Neal Becker wrote: Running update today appeared to hang on restorecon, which was triggered by an update to selinux-policy-targeted IIRC (can't seem to find the log). After running >10minutes (system has SSD and shouldn't take long) I did kill -KILL to it. On reboot everything seems OK. Thoughts? Yes, some selinux updates should come with "warnings" that it may take a long time to complete. This can take quite a bit of time depending on circumstances. I have a slower system and even with an SSD the update took about 30 minutes. I have slower system with a bigger filesystem (ca. 16 TB) on HDDs attached to it, and the update took half a day. Seems to as if SELinux now is relabeling filesystems it did not touch before ;) Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Reasons for staying on a old version ( Re: Prepping for upgrade)
On 3/12/20 8:08 PM, Neil Thompson wrote: Ralf, With all due respect (i.e. none) I don't think your opinion is really necessary here - you've made your feelings clear over the last long while and I have to wonder what the heck you're still doing here if you hate the whole thing so much. Surely there's another distribution that meets your strict requirements better than Fedora? What makes your think, I hate Fedora? What do you think legitimates you misbehave in the you did and to rudely attack and offend me? You can't stand opinions, which diverge from the official Fedora (TM) marketing opinions? I am using Linux for > 25 years, I am using Fedora since its first day and have encountered many of the problems Fedora releases suffer from. Also, I maintain >> 100 packages in Fedora and therefore am pretty sure to know the problems Fedora suffers from. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Reasons for staying on a old version ( Re: Prepping for upgrade)
On 3/12/20 5:10 PM, sixpack13 wrote: On 12.03.20 14:05, Ralf Corsepius wrote: On 3/11/20 11:31 PM, Christopher Marlow wrote: I am just curious why people stay on old versions of Fedora like say FC30 instead of upgrading to 31? Upgrading Fedora is a bit like "installing Win10" updates. That said, I usually wait for a couple of weeks before upgrading, because new Fedora releases tend to suffer from pretty nasty bugs or changes, which often get fixed or at least documented in the early phases after a release. VETO !!! I installed (upgraded to) F31 when it was Beta and used it for daily use. I'm not sure, but I mean I did so with F30. Mere luck ;) Seriously, this is a case of YMMV. Throughout the years, I've experienced all kind upgrades. From entirely flawless to total disaster. Win10 or general Win up{d,gr}ate is a russian roulette with default 5 (!) bullets. Exactly. Though Fedora upgrades in general are not as much russian roulette as MS update, they also carry a non-zero risk. From my experience, the probability of success s significantly higher a couple of weeks later after release, when a Fedora release has seen some "exposure to the wild" and when maintainers had a chance to fix their bugs. Even while Fedora documents their bugs your comparison is invalide. My point was the Russian roulette character. Fedora upgrades share the Russian roulette character with MS-Patch Tuesdays. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Reasons for staying on a old version ( Re: Prepping for upgrade)
On 3/11/20 11:31 PM, Christopher Marlow wrote: I am just curious why people stay on old versions of Fedora like say FC30 instead of upgrading to 31? Upgrading Fedora is a bit like "installing Win10" updates. That said, I usually wait for a couple of weeks before upgrading, because new Fedora releases tend to suffer from pretty nasty bugs or changes, which often get fixed or at least documented in the early phases after a release. I am guessing after a new version of FC comes out, that the previous version is still supported for a limited time? Yes, typically 1/2 year + 1 month. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Fedora 30 nvidia
On 5/5/19 1:26 PM, John Fawcett wrote: Hi Just before I launch into an upgrade to Fedora 30, has anyone had a successful upgrade F29->F30 with nvidia drivers? Yes. An upgrade on an old Core2Duo with an NVidia card and rpmfusion nvidia-akmods worked without major probs (There were minor probs related to the akmods). My graphics card is a bit old by now, latest nvidia driver that supports it is the legacy release 390.116. Mine requires the 304.xx series :) Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: update to F30 seems to have failed
Ralf Corsepius: > That's what I call superflous bureaucracy beyond reason. No, it's merely logical.?? Yes. If it's not available, *and* we're not aware that it's available, it's not released.?? It's as simple as that. It was available, it just had not formally been announced and the tester already stopped working on it. It might just have been that you didn't know about. Trying to say that something is released (but yet not actually available), just because it's said that it will be released in a day or so's time, *is* bureaucratically perverse.?? And just plain nuts. That's the attitude of a bureaucrat: Reject questions for mere formalisms. It's what I hate about Fedora. Some people being involved, seem to be more interested in formalisms and bureaucracy, than being pragmatically helpful and actually do some work on Fedora. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: update to F30 seems to have failed
On 5/3/19 6:51 AM, Tim via users wrote: Allegedly, on or about 2 May 2019, Ed Greshko sent: When is jury selection? I imagine there must be an upcoming trial to address this travesty. :-) My vote is: Until it's officially announced AND released, it ain't a release. Even though we could all be expecting it on a prearranged date, it could be deliberately delayed by circumstances. That's what I call superflous bureaucracy beyond reason. And ignoring issues about imminent releases: Without redirecting people to the correct list, they'll ask in the wrong place (which won't actually do them much good). And without reminding some people, they'll keep on doing it. Face it, posting about user issues with an already finalized release to @test is waste of time and resources. The "testers" already have dropped "their pencils" and are gone. Nobody will listen nor respond. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: update to F30 seems to have failed
On 5/2/19 8:32 AM, Samuel Sieb wrote: On 4/30/19 8:33 PM, Ralf Corsepius wrote: You redirected the OP to test@, at a point in time, when - though fc30 had not been formally announced - fc30 already had been in place on the download servers. That said, you were just behaving bureaucratic and nit-picking, IMHO. Even though it had been technically "released", an upgrade problem right at release is probably better brought to the attention of the QA people. Are you seriously trying to tell us, these folks will look into reported probs, when the release already was rubberstamped "go" several hours before the OPs posting and technically in place on the mirrors? Serious, I find this rediculous. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: GRUB prompt after F29 to F30 upgrade
On 5/1/19 8:24 AM, Chris Murphy wrote: It happens if the most recent 'grub2-install' happened with a Fedora 20 or older GRUB package. e.g. if you installed Fedora 20 clean, then did an OS upgrade every 6-12 months all the way through to Fedora 30, then you'd probably run into this bug. But if at any point from Fedora 21 and newer you ever did a 'grub2-install' you won't hit it. I am not sure if this applies in my case and don't have any possibility to check for it anymore. The pecularity of my system, is it using a gpt formated boot disk though it being a BIOS system (Using a 1M BIOS boot partition). Does Fedora's updater support this case? The installer does. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: update to F30 seems to have failed
On 4/30/19 8:25 PM, Patrick O'Callaghan wrote: On Tue, 2019-04-30 at 18:50 +0200, Ralf Corsepius wrote: On 4/30/19 6:40 PM, Patrick O'Callaghan wrote: On Tue, 2019-04-30 at 17:34 +0200, Ulf Volmer wrote: On 29.04.19 17:51, Patrick O'Callaghan wrote: On Mon, 2019-04-29 at 09:43 -0400, Neal Becker wrote: I just attempted update f29->f30 using sudo dnf system-upgrade download --refresh --releasever=30 F30 has not been released yet. This should be reported on the Fedora Test list, not here. https://fedoramagazine.org/announcing-fedora-30/ That announcement is dated today. My reply was yesterday. Do you think, your reaction was appropriate? In what way was it inappropriate? You redirected the OP to test@, at a point in time, when - though fc30 had not been formally announced - fc30 already had been in place on the download servers. That said, you were just behaving bureaucratic and nit-picking, IMHO. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: update to F30 seems to have failed
On 4/30/19 5:53 PM, Sjoerd Mullender wrote: Now that the release announcement has been done, I can say that upgrading for me (on a VM) also failed. After the final reboot, I just got a grub prompt. So far, the upgrade went fine on 2 out of 3 systems. On the 3rd system, I am also stuck with a grub prompt. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: update to F30 seems to have failed
On 4/30/19 6:40 PM, Patrick O'Callaghan wrote: On Tue, 2019-04-30 at 17:34 +0200, Ulf Volmer wrote: On 29.04.19 17:51, Patrick O'Callaghan wrote: On Mon, 2019-04-29 at 09:43 -0400, Neal Becker wrote: I just attempted update f29->f30 using sudo dnf system-upgrade download --refresh --releasever=30 F30 has not been released yet. This should be reported on the Fedora Test list, not here. https://fedoramagazine.org/announcing-fedora-30/ That announcement is dated today. My reply was yesterday. Do you think, your reaction was appropriate? Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Will RedHat deprecate KDE on Fedora?
On 11/3/18 2:00 AM, William Oliver wrote: I just read this in The Register -- that RH is deprecating KDE in RHEL. As a long time fan of KDE, I'm a bit saddened. Is this planned for Fedora as well? https://www.theregister.co.uk/2018/11/02/rhel_deprecates_kde/ Provided the IBM deal and keeping the thought pronouced by Former Fedora Lead GDK in https://lists.fedoraproject.org/archives/list/council-discuss%40lists.fedoraproject.org/message/WONQLOAXHAV42KXSHZEM4IZQ666YLF3I/ I can not withstand to recommend this RHAT manager to apply for a new job outside of RHAT/IBM. His ROI focused POV is shortsighted and is severely hurting IBM/RHAT's future. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Replacing email list for users with a web forum software called Discourse, what's your opinion?
On 10/22/18 5:27 PM, SternData wrote: I'm OK with a web forum as long as it has an RSS feed. You might not be aware about it, but RSS is effectivly dead, IIRC, because major browsers soon will drop supporting RSS or already dropped it. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Replacing email list for users with a web forum software called Discourse, what's your opinion?
On 10/22/18 4:13 AM, Robbi Nespu wrote: On 10/21/18 5:42 AM, stan wrote: The idea is that it would invigorate the Fedora community by encouraging younger people raised on social media and mobile platforms to contribute. Young people tend to use social media such twitter and facebook but having forum board would be fun too. It easier to access. IMHO, forums are nice to have for occasional use, when not frequently using them and when not closely following. They are not a replacement for high volume mailing lists and when closely following them. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Replacing email list for users with a web forum software called Discourse, what's your opinion?
On 10/20/18 11:42 PM, stan wrote: Hi, There's a big mail thread on fedora-devel about using a web forum software called Discourse instead of mailing lists. The idea is that it would invigorate the Fedora community by encouraging younger people raised on social media and mobile platforms to contribute. One of the targets they have mentioned is the user list, because of its nature of short term question and answer topics. There has been push-back by developers because they have their custom solutions all working great around email lists for doing their work. I'm wondering how the people who regularly use fedora-users mailing list feel about that. Which would you prefer?Mailing lists. I'll likely stop contributing to Fedora if this step takes effect. Unfortunately, it's an either / or proposal, because there is no interface in Discourse for emails from an email list to be put in their forums, though they do have email notification for new web messages. Would you willingly or reluctantly migrate to the new platform? No. IMO, any such forums can not replace mailing lists unless they have an email interface. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Hard drive to sleep
On 10/12/18 4:11 PM, Patrick Dupre wrote: There maybe a good reason why I cannot put this drive at sleep. The /tmp partition is on this disk! Are you using Fedora? For several years, Fedora by default put /tmp on tmpfs. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Hard drive to sleep
On 10/7/18 11:48 PM, Tim via users wrote: Allegedly, on or about 7 October 2018, Ralf Corsepius sent: HW-wise you should check if your drive-hardware is suiteable to be frequently "put to sleep/woken up". Most NAS- or server-class HDDs are not, most desktop/notbook drives are. All the domestic NASs, that I've seen, put their drives to sleep when idle. It can actually be a pain, because they're very slow to wake up. Absolutely. My point actually was, one should check if a HDD actually is designed for this use case. Many NAS- and enterprise/server-class drives are not designed for this use case. Many are designed to be run "24/7" and not for frequent spin ups/downs. One may argue whether this NAS/enterprise/consumer/... classification is more than "manufacturer marketing". Some say so and some say so ;) Anyway, a historic example for such a case is a certain series of WD-Reds (NAS-class HDD) some years ago, which prematurely died when being used with frequent spin up/downs. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: Hard drive to sleep
On 10/7/18 9:29 AM, Patrick Dupre wrote: Hello, Is it safe to put a hard drive on sleep? Depends on your use-case. SW-wise it is pretty save to "put to sleep" not frequently used drives ("data"/"backup" drives). "Putting to sleep" drives hosting system-partitions or partitions hosting your /home/ directory often is not possible. "Putting to sleep" drives, which are hosting seldomly accessed "data" partitions works. HW-wise you should check if your drive-hardware is suiteable to be frequently "put to sleep/woken up". Most NAS- or server-class HDDs are not, most desktop/notbook drives are. I will have to wake it up, later. (I need to minimize the power consumption during a period of time) I am using the approach described in https://wiki.archlinux.org/index.php/hdparm to powerdown drives hosting "data" and "backup" partitions. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
Re: dnf crash
On 03/05/2018 02:55 AM, Eyal Lebedinsky wrote: While it mostly works, this specific command fails: $ dnf provides '*/Droid Sans*' This is consistent, and also happens on a second machine. Both run f27 fully updated. The following does work: $ dnf provides '*/Droid*' But this one crashes: $ dnf provides '*/x y*' I suspect a python problem. Anyone else sees these crashes? Yes, I am seeing these segfaults as well. Interestingly on i386 the segfault is slightly more verbose: # dnf provides '*/Droid Sans*' Last metadata expiration check: 0:00:33 ago on Mon 05 Mar 2018 05:39:27 AM CET. free(): invalid size Aborted (core dumped) Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: can "more" and "less" be used effectively interchangeably?
On 02/18/2018 08:42 AM, Robert P. J. Day wrote: currently perusing some linux courseware i'll be delivering later this month, and i'm fascinated by the early claim that the "more" and "less" commands can be used interchangeably. This does not apply. "more" is the traditional UCB/Berkeley "more" command. It is covered by POSIX and supposed to be portable across POSIX-compatible and *NIXish OS. i don't recall that *ever* being the case, given that more is part of util-linux, and less is a separate package, and less has always had more functionality than more. "less" is a "feature-enhanced" "more". It supports the "more"-CLI syntax, but has additional features added, which are not supported by the traditional "more". Unlike "more", which likely is present on all *NIX/POSIX-compliant OSes (even 20 years old ones), "less" is "optional" or might not be present. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: What is this gibberish?
On 02/04/2018 08:34 AM, Todd Zullinger wrote: The problem wasn't that it was silent. It was that it was a long(ish)-running process that was not suited to run as a scriptlet. It's better done via cron or as it is now as a transient systemd-run service. And does this actually work? I recently was facing situations where this mandb stuff hit midst of shutdown, when all mounted files already where unmounted, delaying shutdowns be some 10-15mins. I haven't investigated, but I was inclined to blame systemd's unreliabily and lack of robustness ;) Anyway, I think the current output is unintentional. I think, the output needs to be more verbose and consider the current output to be non-helpful. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: F26 suddenlty freezes (apparently due nouveau)
On 01/24/2018 12:34 PM, Frédéric wrote: Since 2 days ago, one of my F26 computer freezes (2 freezes in 3 days). I had similar problems on one f27 machine w/ amdgpu. They "magically" healed with today's kernel update. As f26 and f27 kernels are very similar these days and (due to meltdown and spectre) contain fairly "experimental" stuff ... I am inclined to blame the kernel for graphics related issues. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: fc27: nis/ypbind?
On 11/16/2017 12:04 PM, Ulf Volmer wrote: On 16.11.2017 11:32, Ralf Corsepius wrote: On 11/15/2017 06:13 PM, Ralf Corsepius wrote: I have been using yp/nis for ages, but with fc27 things stopped working? Has anybody managed to get ypbind/nis working with fc27? If so, how? To me, yp/nis seems to be in an untested shape and unusable in fc27. i haven't seen an active NIS installation in real life in the last years. You should think about switching to LDAP with or without kerberos. It's my home/office network ;) That's much better tested. Well, I haven't seen a functionally equivalent/usable/maintainable LDAP-Installation anywhere. All implementation lack somewhere ;) Besides this, I don't accept this as excuse for maintainers shipping crap, rsp. them not doing their works - Which apparently is the case wrt. rpcbind/ypbind/ypserv/SELinux (and systemd) Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: fc27: nis/ypbind?
On 11/15/2017 06:13 PM, Ralf Corsepius wrote: Hi, I have been using yp/nis for ages, but with fc27 things stopped working? Has anybody managed to get ypbind/nis working with fc27? If so, how? To me, yp/nis seems to be in an untested shape and unusable in fc27. Some more insights: The culprit is seems to be SELinux. Apparently Fc27's rpcbind/ypserv/ypbind SELinux support is broken Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
fc27: nis/ypbind?
Hi, I have been using yp/nis for ages, but with fc27 things stopped working? Has anybody managed to get ypbind/nis working with fc27? If so, how? To me, yp/nis seems to be in an untested shape and unusable in fc27. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: selinux prevents user to log-in
On 11/09/2017 05:35 AM, Joe Zeff wrote: On 11/08/2017 08:01 PM, Ralf Corsepius wrote: You can't touch /.autorelabel, when you can't log-in ;) You can always boot into rescue mode, touch /.autorelabel and reboot. Actually, I have never used rescue mode and have never found it useful for anything ;) Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: selinux prevents user to log-in
On 11/09/2017 05:26 AM, Dirk Gottschalk wrote: Am 9. November 2017 05:01:44 MEZ schrieb Ralf Corsepius : On 11/08/2017 07:26 PM, Earl A Ramirez wrote: On 8 November 2017 at 12:23, Ralf Corsepius mailto:rc040...@freenet.de>> wrote: On 11/08/2017 04:34 PM, François Patte wrote: Bonjour, I have just upgraded an f23 to f26, keeping the home partition. selinux blocks the user log-in: when I try to log in, lightdm shows my login name, but when I enter my password, the process aborts and I go back to the lightdm login screen Try touch ./autorelabel You can't touch /.autorelabel, when you can't log-in ;) When this happens it means that your SELinux label is incorrect for /etc/shadow Or /etc/ssh ;) You can try to perform restorecron -R /etc/shadow You also can't do this when you can't log-in. Ralf You should still be able to log in as root an a text console True, in most cases this is still possible - But I've also tripped occasions, when even this wasn't possible anymore. Ralfl ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: selinux prevents user to log-in
On 11/08/2017 07:26 PM, Earl A Ramirez wrote: On 8 November 2017 at 12:23, Ralf Corsepius <mailto:rc040...@freenet.de>> wrote: On 11/08/2017 04:34 PM, François Patte wrote: Bonjour, I have just upgraded an f23 to f26, keeping the home partition. selinux blocks the user log-in: when I try to log in, lightdm shows my login name, but when I enter my password, the process aborts and I go back to the lightdm login screen Try touch ./autorelabel You can't touch /.autorelabel, when you can't log-in ;) When this happens it means that your SELinux label is incorrect for /etc/shadow Or /etc/ssh ;) You can try to perform restorecron -R /etc/shadow You also can't do this when you can't log-in. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: selinux prevents user to log-in
On 11/08/2017 04:34 PM, François Patte wrote: Bonjour, I have just upgraded an f23 to f26, keeping the home partition. selinux blocks the user log-in: when I try to log in, lightdm shows my login name, but when I enter my password, the process aborts and I go back to the lightdm login screen Been there, seen that dozens of times. If I disable selinux, I can login. What to do: permanentely disable selinux? Try to initiate relabeling. The probably easiest way to do so is: 1. Reboot 2. Wait for the grub menu to appear, select the kernel you want to boot and press "e" (edit). 3. Scroll down to the line beginning with linux* (e.g. linux16 or linuxefi) and append "selinux=0" (without the '"') to it. 4. Press Ctrl-x (boot) Now the machine should boot with selinux being disabled. When the machine is up, reboot again. SELinux now should be reenabled and the filesystem be automatically relabeled. In the cases I encountered this had helped. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: repos for Fedora 26 i386 (Mate) network install?
On 10/14/2017 11:24 AM, Franta Hanzlík wrote: I want to try Fedora 26 i386 (with my preferred Mate desktop). I prefer network installation with custom kickstart config. I found that install image can be downloaded on: https://download.fedoraproject.org/pub/fedora-secondary/releases/26/Workstation/i386/iso/Fedora-Workstation-netinst-i386-26-1.5.iso But know anyone URLs for usual repos (os, updates, rpmfusion-{free,nonfree}{,-updates}, livna, ... ? I feel as if now, with Fedora 26/i386 moving to fedora-secondary, installation became much more complicated. Why and how? I am aware Fedora's homepage does its best to hide the i386, but just d/l the iso from the URL you mentioned above and try to install it. Unless you have a "really old" machine or are tripping general bugs (I am not using Mate), this should "just work as usual". Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Problem: Cross compile to arm (Raspberry pi) using autotools
On 10/07/2017 01:49 PM, Dirk Gottschalk wrote: Hello, I have a problem cross compiling using autotools for my Raspberry Pi 3. I wrote a server program which runs good with Linux and under windows (compiled with MinGW) on my X86_64 machine. Since there is no configure wrapper for ARM like MinGW has, i use the following configure call: ./configure --host=arm-linux-gnu The Script is failing and the log tells me about bad options "-qverseion" and "-V" in GCC tests. So I removed this Options from the Script by hand. Now it tells me about a missing file called "crt1.o". But i found this file in "/usr/arm-linux-gnu/lib/crt1.o". Well, in general, this indicates a broken path, either in the toolchain or in your compiler call. Try touch tmp.c arm-linux-gnu-gcc -v tmp.c and check the search paths being reported. So, my question is: Did i miss something? Likely ;) Is your target running arm-linux-gnu or something else (e.g. bare metal, a different OS, ...)? Normally, you need a toolchain specialized/dedicated to your target and can't use an arbitrary toolchain. How do I use the cross compiler with autotools? In an ideal world, exactly like you did. However, it's pretty likely you are tripping over bugs in your configure.ac. Also make sure NOT to have set CC, CFLAGS etc in your environment. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: End of i686 Support
On 09/06/2017 01:41 PM, Tim wrote: And, is it still feasible to run the OS on old hardware? Why should it not be? Just because some $DEITY@Redhat has decided you to throw away your old hardware you must do so? Up to fc25 it was technically perfectly possible to run Fedora on a variety of old hardware. Wth f26 things have tightened - f26 runs on some ix86s but doesn't on others. Whether it is feasible for a personal use case, is up to you to decide. I chose Fedora because I run Fedora on all of my machines and because I didn't want to add the personal load to maintain yet another OS. Now RH and FESCO seems to be want to push me to draw a decision - ATM, I am evaluating other distros and am not sure I will continue support Fedora. On my older 32-bit PCs, the notion of running Gnome or KDE is impossible. They're too slow for being fancy with the graphics card. And, some can't even run a modern distro, because they can only take 1 gig of RAM. The minimum specs for running Fedora have crept up and up over the years. Well, don't read too much into the docs. I have been running Fc26 w/ xfce4 on a Pentium III w/ 512MB RAM. No way with Gnome or KDE on this system. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: End of i686 Support
On 09/04/2017 07:18 PM, Andras Simon wrote: 2017-09-04 18:36 GMT+02:00, Ralf Corsepius : [...] Actually, I'd recommend Fedora/RH to drop all other "secondary" archs, because they do not have a community user base. How do you know? - How many arm, s370, ppc or mips machines do you have around? Except of the arm - which some enthusiasts/hobbists may have - probably none. - How many ix86s have been sold? Probably 100s of millions, if not billions. - How many ix86s are still in use? Probably millions, most of them probably running Microsoft Windows. That said, I still have 3 of these around, so far having been running Fedora, but I am now looking for alternative distros because Fedora's leadership apparently is not willing to cater my personal needs. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: End of i686 Support
On 09/04/2017 05:20 PM, Matthew Miller wrote: On Mon, Sep 04, 2017 at 10:32:33AM -0400, Tom Horsley wrote: I'd hope the plan is to do this much the same way as centos. Drop i686 kernels and programs, but continue to provide i686 libraries for legacy 32 bit programs which have no source code so they can continue to run. There are a lot of them out there people use (adobe reader comes to mind). a) I do not agree with this view. b) To my knowledge, Adobe reader already is broken by other packaging bugs in Fedora 26/x86_64. Unless an active maintainer community forms around the i686 kernel and boot process, yes, the above is the plan; we won't drop i686 userspace on x86_64 kernel for... a long time. ... or Red Hat to revert this plan Actually, I'd recommend Fedora/RH to drop all other "secondary" archs, because they do not have a community user base. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: OS for Thinkpad T42??
On 08/19/2017 09:08 PM, j.witvl...@mindef.nl wrote: For the t42 you need to tweak one kernel parameter. T42-CPU does not support pae-extension. Somewhere down the kernel-line they changed the default setting. From default to 'required', thus very old cpu's won't boot anymore, unless you drop this requirement. This would be news to me. Fedora ships pae (package name kernel-pae-*) and non-pae kernels (package name kernel-pae-*). The pae kernels have the requirement you mention, the non-pae kernels should not have this requirement. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: OS for Thinkpad T42??
On 08/18/2017 07:55 PM, Rick Stevens wrote: My guess is yeah, it'll need more memory--regardless of what "modern" OS you manage to install on it. I think even Windows 10 has a minimum of 1GB for a 32-bit environment and 16GB of disk. Fun, fun, fun! Well, installing Fedora 26 on x86ers with less than 1GB RAM definitely is possible. Installation may require some tricks (e.g. manual partitioning to add swap), but it definitely is possible At least, I have Fedora running on a PIII w/ 512 MB RAM: # cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 8 model name : Pentium III (Coppermine) ... # cat /proc/meminfo MemTotal: 505976 kB MemFree: 201332 kB MemAvailable: 388100 kB ... # cat /etc/fedora-release Fedora release 26 (Twenty Six) However, due to the tight RAM constraints, using a DE on it requires a certain amount of "patience". With xfce, the desktop performance is OK for "rare" and "occasional", "light" use (email, browsing, ...), but not for much more - Gnome and KDE are beyond limits ;) Finally, I do not recommend to use Fedora on x86ers anymore, because of Redhat and Fedora's Leaders politics against 32bit. I am considering to switch to Mageia or Debian. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: What is this DNF nonsense?
On 07/25/2017 01:30 PM, Tom Horsley wrote: Almost the very first thing I do on a new install is to erase bash-completion, and it is indeed not installed: tomh> rpm -q bash-completion package bash-completion is not installed So why did the dnf-automatic run I have in cron tell me this last night: Delta RPMs reduced 80.7 MB of updates to 38.6 MB (52.1% saved) The following updates were downloaded on 'tomh': Package Arch Version Repository Size Installing: mod_http2 x86_64 1.10.7-1.fc24 updates 148 k Upgrading: GraphicsMagick x86_64 1.3.26-3.fc24 updates 1.4 M GraphicsMagick-doc noarch 1.3.26-3.fc24 updates 973 k httpd x86_64 2.4.27-3.fc24 updates 1.3 M httpd-filesystemnoarch 2.4.27-3.fc24 updates 27 k httpd-manualnoarch 2.4.27-3.fc24 updates 2.3 M httpd-tools x86_64 2.4.27-3.fc24 updates 89 k icedtea-web x86_64 1.7-1.fc24updates 1.8 M java-1.8.0-openjdk x86_64 1:1.8.0.141-1.b16.fc24updates 233 k java-1.8.0-openjdk-develx86_64 1:1.8.0.141-1.b16.fc24updates 9.8 M java-1.8.0-openjdk-headless x86_64 1:1.8.0.141-1.b16.fc24updates 32 M librsvg2x86_64 2.40.18-1.fc24updates 133 k librsvg2-devel x86_64 2.40.18-1.fc24updates 52 k qt5-qtwebengine x86_64 5.6.3-0.1.20170712gitee719ad313e564.fc24 updates 31 M Skipping packages with conflicts: (add '--best --allowerasing' to command line to force their upgrade): bash-completion noarch 1:2.4-1.fc24 updates 269 k Do you have "install_weak_deps=false" in your /etc/dnf/dnf.conf? These bewildering, bizarre and confusing "conflict messages" are dnf's poor way to tell it did not install a packages with weak dependency :( Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Green screen error in VLC
On 07/13/2017 01:56 PM, Paul Smith wrote: On Thu, Jul 13, 2017 at 12:44 PM, Ralf Corsepius wrote: This happens after upgrading from F25 to F26, when playing a video -- no image; only a green screen. Any ideas? Which GPU do you have and what is your setting in vlc's "Tools->Preferences->Video->Output"? Since having upgraded to F26[1], I facing similar issues as you describe them on 2 Haswell's IGPUs. Playing with "Output", I found the only working one seems to be "XVideo output (XCB)" while all the rest - comprising "Automatic" - seem dysfunctional. Ralf [1] I did not have these issue on the same machines with FC25. One of these machine is in a dual boot config with F25, so I can try FC25 (when time permits). Thanks, Ralf. By setting "Output" to "XVideo output (XCB)", instead of "Automatic", did the trick. But should this be considered a bug? IMO, yes, definitely. The problem however is, it's hard to tell which component (package) is the actual cause. vlc is involved but from what I see from the error messages vlc emits, there are more bugs involved underneath. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Green screen error in VLC
On 07/13/2017 01:26 PM, Paul Smith wrote: Dear All, This happens after upgrading from F25 to F26, when playing a video -- no image; only a green screen. Any ideas? Which GPU do you have and what is your setting in vlc's "Tools->Preferences->Video->Output"? Since having upgraded to F26[1], I facing similar issues as you describe them on 2 Haswell's IGPUs. Playing with "Output", I found the only working one seems to be "XVideo output (XCB)" while all the rest - comprising "Automatic" - seem dysfunctional. Ralf [1] I did not have these issue on the same machines with FC25. One of these machine is in a dual boot config with F25, so I can try FC25 (when time permits). ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: /var/cache
On 07/05/2017 10:16 AM, Patrick Dupre wrote: but it is even worst for /var/cache: 14338744 I gives the largest sub directories: How can I clean this? Files under /var/cache are supposed to be automatically regenerated by the programs which create them (Hence the name "cache"). I.e. in theory, you can erase everything under /var/cache whenever you want. If this breaks something, this would qualify as a bug. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: FedoraPackages
On 02/14/2017 06:26 PM, InvalidPath wrote: Probably a dumb question but if you see a package release pushed to stable.. does that mean it’s available in the Fedora repos? Yes. Reality is bit more complicated, but somewhat oversimplfied, that's what it means. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: SUB# connectors -
On 02/06/2017 04:53 PM, Bob Goodwin wrote: I have an MSI motherboard with USB3 connectors on the back panel that I have never been able to use because the connectors are different and nothing I have fits them.They are similar to the usual USB 2 connectors but have a blue insert. Errm, USB-3 connectors are supposed to be mechanically and electrically backward compatible to USB-2. I.e. a USB-2 device's plug (white or black) should (mechanically) fit into a board's USB-3 socket (blue) and "just work". Normally, USB-connectors on backplanes are "Type-A". I would like to order adapter cables, whatever but a lot of googling hasn't helped ... I've looked for this before without success. You don't need any special cables or adaptors. May-be you're confused about the Mini-/Micro-/A-/B-form-factor variations? Can anyone help me with the proper terminology to order something that fits? c.f. https://en.wikipedia.org/wiki/USB Scroll down to "Host and device interface receptacles". There is a socket/plug compatibility matrix/table. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: RANT: installing fedora is now a real punishment!
On 02/01/2017 02:34 PM, François Patte wrote: Install of f25 was (almost) easy *but* I am unable to login! I created a user with a password but when I want to login system claims that the password is incorrect. OK maybe I made a mistake, so I try to login as root from the console : password is incorrect too! I've seen this happening, when permissions and/or ownership of ssh-related directories were too open or incorrect[1]. To remedy this, try to boot from another medium (dvd/usb), mount your / (and /home) and check permissions/ownership of /etc/ssh/* and /home//.ssh. What I also experienced, was being unable to initially log-in with SELinux enabled, seemingly because SELinux was broken after upgrades. There, I first had to boot with SELinux disabled (Adding selinux=0 to the grub-prompt) to get into the machine, issue a "touch /.autorelabel" and reboot. Ralf [1] Seems to me as if previous Fedora versions allowed to set them to what ssh nowadays considers "too open" and upgrades carried these over. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: no shutdown at all in fedora 24 Again!
On 01/22/2017 02:26 AM, Chris Murphy wrote: I think you're going to have to put logs up somewhere and configuration information somewhere to get an answer. I can tell you that mdadm raid and lvm being used do not themselves result in an inability to shutdown Fedora. Fedora by default uses LVM, and mdadm raid0/1/5 tests are done on pre-release versions of Fedora; if any of that resulted in an inability to restart or shutdown, it'd likely be a release blocking bug. So the mere fact systemd itself does not directly support md raid or lvm doesn't really translate into a cause of your problem. I am also facing these shutdown probs and have come to the conclusion, that there is "no single cause", but that they are the result of a combination of several causes, probably related to systemd, NetworkManager, SELinux and other, networking related packages. Isolating them seems to be difficult, because almost each update changes and breaks something else. E.g. right now, rpcbind/nfs/yp/nis's integration into systemd seems to be "mash" in fc25 - It has "bricked" one of my machines, today! Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Most system update requires system reboot or session restart
On 12/21/2016 02:00 AM, Kevin Fenzi wrote: On Tue, 20 Dec 2016 15:39:08 -0800 Joe Zeff wrote: On 12/20/2016 03:27 PM, Matthew Miller wrote: Depends on which part. You can use dnf repoquery to list duplicates, leaf packages (now "unneeded", which I don't think is an improvement in clarity, but whatever), orphans ("extras", and ditto), unsatisfied deps (formerly "problems", now "unsatisfied" — that one is an improvement). The cleandupes functionality is now "dnf remove --duplicates". That's good to know. Somehow, upgrades seem to hang on this system (but not my laptop) leaving me unable to start a GUI, and with so many dupes that I'd have to get a list of them (package-cleanup --dupes | grep fcXX> dupes.txt with XX representing the older disty because it fcXX> listed both sets) and then clean them up a few at a time by hand. Not the most pleasant task, but doable as long as I've got a simple way to get that list. FYI, If you are looking for a particular dnf command that does something yum used to do the first place to look is 'man yum2dnf' (or https://dnf.readthedocs.io/en/latest/cli_vs_yum.html ) While we're at it: No idea, what "dnf repoquery --unneeded" does, but it by no means is an equivalent to "package-cleanup --leaves": # package-cleanup --leaves | wc -l ... 40 # dnf repoquery --unneeded [nothing] It's one of the reasons I still have package-cleanup installed. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Kernel not configured for semaphores
On 12/12/2016 07:24 AM, Ed Greshko wrote: On 12/12/16 13:35, Frédéric Bron wrote: Hi, Each time I turn off Fedora 24, I get the following message about 50 times printed in the console: Kernel not configured for semaphores (System V IPC). Not using udev synchronisation code. device-mapper: remove ioctl on fedora_alpha-root failed: Device or resource busy Command failed What does that mean? > It means you should follow this bugzilla https://bugzilla.redhat.com/show_bug.cgi?id=1359352 I am inclined to believe this bug might not be related to the OP's problem - At least not the incarnation, I am facing ;) What I am seeing on several machines, is larger series of the error message above rushing through very rapidly (<< 1sec) upon each shutdown on several machine, but so far I have not been experiencing shutdown hangers. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Fedora25: SDDM not showing NIS accounts
On 12/05/2016 09:12 AM, Samuel Sieb wrote: On 12/04/2016 11:54 PM, Terry Barnaby wrote: On 05/12/16 06:43, Samuel Sieb wrote: On 12/04/2016 07:42 AM, Terry Barnaby wrote: In F25 the SDDM no longer shows a list of users when NIS (ypbind) is enabled and also it does not default to the last user logged in. kdm appears to be the same (with UserList=true in /etc/kde/kdm/kdmrc). Are you expecting all the users to be listed or just the ones that have used that computer before? I would have assumed all users (uid 1000 up). Displaying all users might be applicable in very small networks with very small numbers of users, but would not be behavior helpful in larger networks with larger numbers (100s/1000s) of users/accounts. It might show all local users, but at least with LDAP, it only shows users that have actually logged in at some point. The AccountService is involved in that. I don't know if it's possible for the login manager to find all the users from an LDAP source, but if it could, that would most likely be too large a list to be displaying like that anyway. I don't know if NIS works differently, but I expect it would have the same issues. I don't know much about LDAP, but in NIS-enabled networks, all non-local users/accounts usually are available at once. Ralf ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
fc25 ypbind/nis fails
Hi, has anybody managed to get ypbind working with f25? If so, how? ypbind used to work with fc25 prereleases[1,2], but with the post fc25-release update to ypbind-1.38-6.fc26.x86_64, ypbind fails at system startup/when booting. Ralf [1] ypbind-1.38-5.fc24.x86_64 [2] nss_nis is installed ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: Weird window behavior
On 08/06/2016 01:33 AM, Tom Horsley wrote: On Fri, 5 Aug 2016 16:04:34 -0700 Rick Stevens wrote: Another example: If I edit a file and change something in vi in an Xterm...the display flickers between what I just edited and what it looked like before the change. Again, clicking in another window makes it stop. I am facing similar issues with f24 on a pure Intel machine. So far I have not seen these issues on machines with NVidia GPUs or with NVidia+Intel-GPU (This doesn't mean much, because the Intel-box is my primary machine, while the others are much less frequently used). My wild guess on the cause would be a low-level graphics related SW-component (Kernel, xorg-X11, libva-intel*, ...). Unfortunately, though I have been trying various things, I have not been able to isolate the cause. However, I feel, things have gradually worsened, which could indicate there might be several bugs interacting. Sounds somewhat like this bug which went away for me when I installed an nvidia card and no longer used the motherboard's intel graphics support: This is not an option to me ;) Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://lists.fedoraproject.org/admin/lists/users@lists.fedoraproject.org Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: old packages
On 08/01/2016 09:54 AM, Patrick Dupre wrote: What are the alternatives ? One would have to look into each of these packages individually to answer this. For example to system-config-lvm ? system-config-lvm still works (Simply rebuild it locally). Actually, I never understood, why RH abandoned it. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://lists.fedoraproject.org/admin/lists/users@lists.fedoraproject.org Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Fedora 24 - Beta installation problem to Netbook
On 05/17/2016 08:58 AM, Prof. Server Acim wrote: Hello, I have a CREA Netbook. I am having problems about installing Fedora 24 Beta. These are the steps and the results of my things that I'd done. Is this an Atom N270 based netbook? If so, I'd guess your are facing https://bugzilla.redhat.com/show_bug.cgi?id=1307033 rsp. https://bugzilla.redhat.com/show_bug.cgi?id=1303860 AFAICT, this bug affects all kernels >= 4.5 on Atom N270 i945GSE based machines. ATM, this means, Fedora > 23 and - due to the fact kernel-4.5.x is in the process of being pushed to f23 - Fedora 23 will likely be unusable on at least some of these machines in near future. I am also not sure that am I asking the right question to the right email group. Perhaps, I may have to ask these questions to the "testing group". Like others wrote before, reporting your issue to "testing" would have been more appropriate at this point in time. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: http://lists.fedoraproject.org/admin/lists/users@lists.fedoraproject.org Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: DNF is a time waster
On 03/30/2016 09:11 AM, Jan Zelený wrote: On 30. 3. 2016 at 06:35:19, Ralf Corsepius wrote: On 03/29/2016 07:36 PM, CLOSE Dave wrote: If I run something like, "dnf -y install ", and is already installed, DNF gives a response like the following. Last metadata expiration check performed 0:00:02 ago on ... Package is already installed, skipping. ... (long delay, sometimes several minutes) ... Dependencies resolved. Nothing to do. Complete! That is, it works but it takes an unreasonably long time to do so. How can it be that, DNF knows the package is already installed and /still/ takes such a long time to decide that dependencies are ok? There aren't any dependencies to resolve! I've seen and am seeing this happening many times, but I've given up complaining about it, because I feel feedback on the numerous defects dnf has, is politically unwanted. Feedback is always welcome, as long as it's constructive and both sides want to find mutually acceptable solution. You certainly know that I am tired of hearing sentences like these. To me this, are all shallow and weak excuses. As a German proverb says: "The bucket goes to the well until it breaks". Now the bucket is broken (read: my patience with dnf and the folks behind it is finally finished) I am expecting Red Hat to finally take consequence on dnf and to stop shipping broken banana SW, such as dnf Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: DNF is a time waster
On 03/29/2016 07:36 PM, CLOSE Dave wrote: If I run something like, "dnf -y install ", and is already installed, DNF gives a response like the following. Last metadata expiration check performed 0:00:02 ago on ... Package is already installed, skipping. ... (long delay, sometimes several minutes) ... Dependencies resolved. Nothing to do. Complete! That is, it works but it takes an unreasonably long time to do so. How can it be that, DNF knows the package is already installed and /still/ takes such a long time to decide that dependencies are ok? There aren't any dependencies to resolve! I've seen and am seeing this happening many times, but I've given up complaining about it, because I feel feedback on the numerous defects dnf has, is politically unwanted. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: conflict installing i686 version of "qt" on x86_64
On 03/15/2016 05:30 PM, Rex Dieter wrote: Ralf Corsepius wrote: On 03/14/2016 07:18 PM, Christian Groessler wrote: Hi, Seems to be a problem that 32 and 64 versions of the krb5-libs package claim/provide the same file. Any idea? Clearly a packaging bug. File a BZ against these packages. It's not a packaging bug. Yeah, I see. You are right - Sorry for that. Try 'dnf update' first, I'm betting you have an older out-of-date krb5- libs-1.14-9.fc23.x86_64 currently installed, updated to latest 1.14.1-1.fc23 should fix it. It's a conflict been 2 versions (1.14-9.fc23 and 1.14.1-1.fc23), likely due to a corrupt rpmdb, a defect in dnf or dnf/rpm getting killed during rpm transistions I've seen all such cases in recent times, esp. the last case (seems to me as if something inside of rpms occasionally is killing something vital midst of "dnf update"s.) Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: conflict installing i686 version of "qt" on x86_64
On 03/14/2016 07:18 PM, Christian Groessler wrote: Hi, I want to install the 32-bit version of QT on a x86_64 system (for Skype, their Linux version seems to be quite outdated). I'm getting this error: -- # dnf install qt.i686 Last metadata expiration check: 0:10:48 ago on Mon Mar 14 19:03:00 2016. Dependencies resolved. Package Arch Version Repository Size Installing: krb5-libs i686 1.14.1-1.fc23 updates858 k openssl-libs i686 1:1.0.2g-2.fc23 updates998 k qt i686 1:4.8.7-5.fc23 updates 5.0 M Transaction Summary Install 3 Packages Total size: 6.8 M Installed size: 23 M Is this ok [y/N]: y Downloading Packages: [SKIPPED] qt-4.8.7-5.fc23.i686.rpm: Already downloaded [SKIPPED] openssl-libs-1.0.2g-2.fc23.i686.rpm: Already downloaded [SKIPPED] krb5-libs-1.14.1-1.fc23.i686.rpm: Already downloaded Running transaction check Transaction check succeeded. Running transaction test The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: Transaction check error: file /usr/share/doc/krb5-libs/NOTICE from install of krb5-libs-1.14.1-1.fc23.i686 conflicts with file from package krb5-libs-1.14-9.fc23.x86_64 file /usr/share/doc/krb5-libs/README from install of krb5-libs-1.14.1-1.fc23.i686 conflicts with file from package krb5-libs-1.14-9.fc23.x86_64 file /usr/share/licenses/krb5-libs/LICENSE from install of krb5-libs-1.14.1-1.fc23.i686 conflicts with file from package krb5-libs-1.14-9.fc23.x86_64 file /usr/share/man/man5/k5identity.5.gz from install of krb5-libs-1.14.1-1.fc23.i686 conflicts with file from package krb5-libs-1.14-9.fc23.x86_64 file /usr/share/man/man5/k5login.5.gz from install of krb5-libs-1.14.1-1.fc23.i686 conflicts with file from package krb5-libs-1.14-9.fc23.x86_64 file /usr/share/man/man5/krb5.conf.5.gz from install of krb5-libs-1.14.1-1.fc23.i686 conflicts with file from package krb5-libs-1.14-9.fc23.x86_64 Error Summary - # -- Seems to be a problem that 32 and 64 versions of the krb5-libs package claim/provide the same file. Any idea? Clearly a packaging bug. File a BZ against these packages. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: smplayer volume control
On 01/25/2016 11:35 PM, Andre Robatino wrote: Andre Robatino fedoraproject.org> writes: Thanks for confirming. Filed https://bugzilla.rpmfusion.org/show_bug.cgi?id=3955 . Thanks. Apparently it's GNOME-specific. I don't think so - I am using xfce ;) With smplayer and avidemux_qt4, I am observing what jd1008 described. Audio "Volume control" is a circle, may be a slider with very small length. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: smplayer volume control
On 01/25/2016 03:47 PM, Andre Robatino wrote: For a long time now, the smplayer (from RPMFusion) volume control is vertical, instead of horizontal, with a negligible height, so it's essentially useless. This is still true with the latest version smplayer-16.1.0-1.fc23, and even after deleting ~/.config/smplayer. Do all Fedora users see this, and can it be fixed? I am not using smplayer at a regular basis, but your posting caught my attention because am facing a similar issue with avidemux ;) I seem to be facing the same issue as you with smplayer. Therefore I would suggest you to file a bug against smplayer rpmfusion's bugzilla. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: NFS URLs in Nautilus on Fedora 22 stopped working
On 01/22/2016 08:26 PM, Ranbir wrote: On Thu, 2016-01-21 at 15:01 -0800, Rick Stevens wrote: Is your NFS server doing NFSV3 or NFSV4? "showmount -e" doesn't work if the server is NFSV4 and I suspect that's what Nautilus is doing (or something like it). No, that's not it: I'm using nfsv3. This could explain your problems. Besides, the mounts worked perfectly before in Nautilus. It's just that now Nautilus isn't recognizing nfs URLs at all. Some time around fc21/fc22, the defaults for mount-type for nfs mounts was changed to nfsv4. I.e. the meaning of file type "nfs" in calls to mount and in /etc/fstab was changed from "nfs"=="nfsv3" to "nfs"=="nfsv4". I also recall, their were some mistakes/oversights in this transitions, which were fixed later on. I.e. if your server is nfsv3, you should check your nfs-related config-files in /etc/. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: flash plugin
On 12/23/2015 08:48 AM, Brandon Vincent wrote: On Wed, Dec 23, 2015 at 12:18 AM, Antonio M wrote: is anybody experiencing this issue?? Tnx The last version of Adobe Flash Player that will work on Firefox for Linux is 11.2. Security updates will be provided until 2017. The major and minor version numbers have not changed for the Linux plugin since 2012. Since work continues on the Windows and OS X versions of Flash, the version numbers are much higher (currently at 20). It is poor practice in my opinion to blacklist/whitelist browser and plugin versions. It only causes trouble down the road. The best solution if Flash Player is a necessity to you is to install Google Chrome or Chromium with the Pepper Flash plugin. IMO, the best solution is to contact this site's admin and tell them their site is broken and that they should take it off-line. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Firefox does not play Rutube videos
On 12/18/2015 09:56 PM, Paul Smith wrote: Interestingly, the Adobe website asserts that the current (and last) version of the flash player is Version 11.2.202.554, but the rpm that the website delivers is adobe-release-x86_64-1.0-1, built in 2011. adobe-release-x86_64-1.0-1 contains their repository's yum/dnf /etc/yum.repos.d/*.repo for x86_64. To get flash going you need to install the flash-plugin from their repository via yum. And indeed, they ship 11.2.202.554: # rpm -q adobe-release-x86_64 flash-plugin adobe-release-x86_64-1.0-1.noarch flash-plugin-11.2.202.554-release.x86_64 cf. http://linuxconfig.org/how-to-install-adobe-flash-player-on-fedora-linux-with-firefox Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: X server dies on startup with kernel-4.2.7-300.fc23.x86_64
On 12/19/2015 01:44 AM, Ed Greshko wrote: On 12/18/15 20:14, Patrick O'Callaghan wrote: After rebooting I get the login screen but Plasma segfaults and Gnome gives an oops and sad face so I've had to fall back to LxQT until I can figure out what's going on. FWIW, I can confirm that akmod-nvidia-358.16-1.fc23.x86_64 breaks my system with the GeForce 8600M GS card (a rather old card in a laptop). Falling back to akmod-nvidia-340xx.x86_64 got things working again. You may want to consider giving that a try after checking and confirming 340xx is supposed to support your card. That's to be expected. One cannot blindly install the latest "akmod-nvidia*". You need to find out which of them is supposed to support your GPU on [1] and choose the corresponding *kmod-nvidia package. According to this page the *-358.* series doesn't support the "GeForce 8600M GS" [2]. I have an older desktop with a "GeForce 8600 GT", where the same applies. Ralf [1] http://www.nvidia.com/object/unix.html [2] Check the "supported products" tab on http://www.nvidia.com/Download/driverResults.aspx/95165/en-us -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Why was YUM removed
On 12/09/2015 07:33 PM, jd1008 wrote: On 12/09/2015 11:27 AM, Joe Zeff wrote: On 12/09/2015 03:21 AM, Tim wrote: As acronyms go, I think it's an odd one, too. DNF in sporting parlance means did not finish. To other people, it might mean do not f**k. Why couldn't you spell out fork properly. Or were you referring to fsck? Seems to me that the better question to have been asked by the OP would have been: What was WRONG with yum? and: What does dnf fix that was broken in yum? My view: The only thing that was wrong with yum, was it being work-in-progress, when its maintainer passed away. Later somebody fell into the common trap of believing a rewrite was superior/easier than gradual improvements. IMO, dnf once more proved this consideration wrong. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: MP4 video on F23
On 12/09/2015 01:36 PM, Patrick O'Callaghan wrote: On Wed, 2015-12-09 at 13:17 +0100, Ralf Corsepius wrote: On 12/09/2015 01:02 PM, Patrick O'Callaghan wrote: On Wed, 2015-12-09 at 12:52 +0100, Ralf Corsepius wrote: On 12/09/2015 12:29 PM, Patrick O'Callaghan wrote: BTW, HandBrake was also failing with the same error, until an update last night fixed it. It no longer tries to load the library in question. HandBrake is neither available from rpmfusion nor fedora. True, it's from a specialized repo: handbrake.fr I would be news to me, handbrake.fr is shipping fedora packages. Where did you get this package from? From negativo17.org. They have a Fedora repo. I assume they compiled HB from sources. It seems to run fine on Fedora. Then it's worth a try to 1. disable this repo, 2. run "dnf distro-sync --refresh --best" and/or run "package-cleanup --orphans" 3. eliminate all orphan package. If this solves your problems, we know the cause. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: MP4 video on F23
On 12/09/2015 01:16 PM, Michael Schwendt wrote: No idea yet where that lib comes from: Review Request: vo-aacenc - VisualOn AAC encoder library https://bugzilla.rpmfusion.org/show_bug.cgi?id=1742 Huh? Else there seem to be packages for OpenSUSE and a few other 3rd party repos. Debian also has it: https://packages.debian.org/en/source/jessie/vo-aacenc Make me think Patrick might have some "blindly" debian-"ported" packages installed. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: MP4 video on F23
On 12/09/2015 01:02 PM, Patrick O'Callaghan wrote: On Wed, 2015-12-09 at 12:52 +0100, Ralf Corsepius wrote: On 12/09/2015 12:29 PM, Patrick O'Callaghan wrote: BTW, HandBrake was also failing with the same error, until an update last night fixed it. It no longer tries to load the library in question. HandBrake is neither available from rpmfusion nor fedora. True, it's from a specialized repo: handbrake.fr I would be news to me, handbrake.fr is shipping fedora packages. Where did you get this package from? I.e. you have other repos in addition to these active and now likely are facing compatiblity issues between them. I see no indication of that. Compatibility issues should show up when updating, but they don't. I run "dnf update" update every day. That said, try "dnf distro-sync --refresh --best" and show use the result. I would expect it to report conflicts. Tried that. No errors (apart from wanting to update xorg-x11-*, which currently I have blocked because of Nvidia issues). Then I can't help you. As Michael wrote before, there is no libvo-aacenc.so.0 anywhere in Fedora nor rpmfusion, nor does any of the packages/program you claim are requiring this library, require this library in Fedora/rpmfusion. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: MP4 video on F23
On 12/09/2015 12:29 PM, Patrick O'Callaghan wrote: BTW, HandBrake was also failing with the same error, until an update last night fixed it. It no longer tries to load the library in question. HandBrake is neither available from rpmfusion nor fedora. I.e. you have other repos in addition to these active and now likely are facing compatiblity issues between them. That said, try "dnf distro-sync --refresh --best" and show use the result. I would expect it to report conflicts. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: MP4 video on F23
On 12/08/2015 06:35 PM, Sylvia Sánchez wrote: A fresh install of Fedora. Sorry! Sorry, but this is bad advice - Linux/Fedora is not Windows. Reinstalling the distro because some arbitrary program (here: vlc) or a library (here: x264) doesn't work/malfunction, is *never* required. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Installing i686 on x86_64 with DNF
On 11/19/2015 03:23 PM, Richard Shaw wrote: Am I'm missing something obvious? I need to install nosync 32bit for mock but I had to remove it before dnf system-upgrade would stop complaining. Now that I'm upgraded I tried "dnf install nosync.i686" does not find the package. What magic incantation is required to install i686 packages on x86_64 wit dnf? Nothing. "dnf install nosync.i686" should do the job, but AFAIS, nosync.i686 is missing from the x86_64 repository. No idea why. I would suggest you to BZ a bug against nosync and ask nosync's maintainer to contact have rel-eng to have nosync.i686 added to the x86_64 repository. If you really need it "RSN", you can resort to manually download dl.fedoraproject.org/pub/fedora/linux/development/rawhide/i386/os/Packages/n/nosync-1.0-4.fc24.i686.rpm and to install it. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: How much cache does keepcache keeps?
On 11/17/2015 08:44 AM, Porfirio Andres Paiz Carrasco wrote: 2015-11-16 23:41 GMT-06:00 Ralf Corsepius : On 11/17/2015 05:28 AM, Sudhir Khanger wrote: Hi, I learned it the hard way that dnf undo/rollback will fail if packages to rollback to are missing. I would recommend not to waste time on rollbacks, because a package based rollback will only work when a package's installation is non-stateful. This applies to many "trivial packages", but does not apply in general. Rollbacks works the same like downgrading?, Is the same task? I am not sufficiently familiar with dnf's internals, but AFAICT, yes. Downgrading is important when, for example you need to use xorg 1.17 instead of xorg 1.18 in order to install a nvidia privative driver because nvidia does not works with the new xorg version. The same consideration here: It's mere luck, if a downgrade works. It's just that most packages are non-stateful, which lets downgrading most packages succeed. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: How much cache does keepcache keeps?
On 11/17/2015 05:28 AM, Sudhir Khanger wrote: Hi, I learned it the hard way that dnf undo/rollback will fail if packages to rollback to are missing. I would recommend not to waste time on rollbacks, because a package based rollback will only work when a package's installation is non-stateful. This applies to many "trivial packages", but does not apply in general. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: dnf gets stuck on poor connections
On 11/13/2015 12:53 PM, Sylvia Sánchez wrote: Mmmhh... My connection isn't very reliable either but dnf doesn't get stuck. Maybe you should update packages in small bunches instead of everything altogether. I am not on a poor connection, either, but I am occasionally experiencing this problem as well. In most cases the cause seems to be dnf's mirror selection to prefer broken repos and/or poorly accessible mirrors, but I have also seen cases, when dnf hung without any feedback for hours. My recommendation: Try yum, dnf is not ready for production use. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Is upgrading to fedora 23 safe?
On 11/04/2015 07:01 PM, Sergiu Mihuleac wrote: Hi! I want to know how many did the upgrade and worked perfectly vs how many had problems and what where those problems. IMO, upgrading is not safe, ATM. I recommend to wait for a couple of weeks. Issues, I have experienced so far: 1. packages with broken upgrade paths. 2. rpmfusion for fc23 isn't complete yet. 3. incompatibilities between the x-server and closed source x-drivers. I am confident to believe 1 + 2 will be resolved in not too distant future, but if you depend on 3, you'll have to wait until those vendors will update their packages or to get away without these x-drivers. Conversely, if you don't depend on proprietary closed-source x11-drivers, and don't depend too much on rpmfusion, chances are good, upgrades will be successful. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: rpmbuild error on F23
On 11/04/2015 03:20 PM, Ranjan Maitra wrote: How does one create debuginfo packages (is there an easy/automated way?) They normally are being generated automatically as by-product of building rpms. However, you a) must not to use "%global debug_package %{nil}" (this disables building debuginfos) b) need to verify the compiled parts of the package honor ${RPM_OPT_FLAGS} rsp. %{optflags}. c) need to verify not to be stripping binaries (binaries must not be stripped. rpm strips binaries itself). However, there are many more possible causes, which renders it hard to help you without having seen the actual src.rpm. From what I see from the *.spec in http://paste.fedoraproject.org/286698/44661012/ you are using "make install-strip". Likely, c) is the cause. => Remove "%global debug_package %{nil}" and try to use "make install" instead of "make install-strip". Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: updated f22->f23 using dnf system-upgrade, no significant issues
On 11/04/2015 11:43 AM, Ed Greshko wrote: Some folks having difficulties may be better served by making sure they have both dnf-plugin-system-upgrade-0.7.0-1 and python2-dnf-plugin-system-upgrade-0.7.0-1 installed. This doesn't change much in my case. The sad truth is the f23 repos currently are inconsistent. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: updated f22->f23 using dnf system-upgrade, no significant issues
On 11/03/2015 03:42 PM, Neal Becker wrote: A boring update Not for me: # dnf system-upgrade --releasever 23 download ... Skipping packages with broken dependencies: python-pexpect What's interesting about python-pexpect is its state in the repos: dl.fedoraproject.org/pub/fedora/linux/updates/23/SRPMS/p/python-pexpect-4.0.1-4.fc23.src.rpm dl.fedoraproject.org/pub/fedora/linux/development/23/source/SRPMS/p/python-pexpect-3.1-4.fc23.src.rpm dl.fedoraproject.org/pub/fedora/linux/development/23/i386/os/Packages/p/python-pexpect-3.1-4.fc23.noarch.rpm dl.fedoraproject.org/pub/fedora/linux/development/23/x86_64/os/Packages/p/python-pexpect-3.1-4.fc23.noarch.rpm In other words, the 23 update repos in inconsistent shape. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Upgrade advice
On 11/02/2015 02:47 PM, Jeffrey Ross wrote: I think I'm going to just bite the bullet and do a fresh install while preserving /home, I recommend to also backup /etc. Otherwise you'd have additional efforts with restoring passwd/accounts, ssl-keys, etc. I'll have to look through the current install guidelines to make sure I'm not breaking some cardinal rule but at this point I'm planning on /boot - Raid1 / - Raid1 /var - Raid1 /home - Raid1 swap - Raid1 What is the current recommendation for swap? I'm guessing it needs to be at least the same size as physical memory. Primarily depends upon your actual amount of memory. On "big RAM" systems you probably don't need any on "small RAM" systems you probably will want some. How to define "small/big" depends largely upon the footprint your typical style of usage imposes. Years ago unix upon a system crash would start writing the crash file at the beginning of the swap partition and keep going until it was done, if swap was smaller than physical memory you had a problem as the system would happily write over the next partition. IIRC, in the old days, the rule was twice the amount of RAM, with the reason being "suspend to disk". As diskspace is cheap these days, I am still applying this rule. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: upgrade to 23 when it comes out
On 10/31/2015 07:35 AM, Porfirio Andres Paiz Carrasco wrote: 2015-10-31 0:16 GMT-06:00 William Biggs : I installed xface f22 64 bit then installed Cinnamon Desktop . When 23 comes out if I upgrade using fedup . I think that what is is called . Will I get the Cinnamon Desktop . or the xface desktop since it was the core install ? I think fedup is not more available. It still is available, but has been re-written as a wrapper around dnf-plugin-system-upgrade: # rpm -qf /usr/bin/fedup dnf-plugin-system-upgrade-0.4.1-1.fc23.noarch Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: strange issues with f22
On 10/14/2015 09:23 PM, Paolo Galtieri wrote: Folks, I have seen several issues with f22 on one of my systems. Issue #1: I tell the system to shutdown, and it doesn't. I see the message "Powering off" on the console and then the system restarts. It does not power off. This doesn't happen all the time, but I have seen it happen on more than one occasion. This resembles very much to these bugs: https://bugzilla.redhat.com/show_bug.cgi?id=1189107 https://bugzilla.redhat.com/show_bug.cgi?id=1257131 Sounds like may try to - add xhci_hcd.quirks=262144 to you kernel boot parameters (in grub2-efi.cfg) - build a custom kernel with the patch from https://bugzilla.redhat.com/show_bug.cgi?id=1257131#c15 applied Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: fedup leaves me stuck between 21 and 22
On 10/06/2015 07:18 PM, sean darcy wrote: running updated 21. fedup --network 22 Preparation seemed to go well: [ 191.911] (II) fedup:() /usr/bin/fedup exiting cleanly at Tue Oct 6 12:26:25 2015 Rebooted. Died somewhere in upgrading. Is there any log of the upgrade ? No new kernel installed. Reboots into 21 kernel. Lots of dupes. yum upgrade just give 21 updates. Lot of dupes means something went wrong during the upgrade, probably has corrupted the rpmdb. You need to clean up the dupes and try to bring things in consistent shape again. Try using "yum/dnf distro-sync", "package-cleanup --dupes" and "rpm". Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: firefox?
On 09/27/2015 10:32 PM, Ed Greshko wrote: On 09/27/2015 10:10 PM, Ralf Corsepius wrote: On 09/27/2015 03:24 PM, Andre Robatino wrote: Paul Cartwright gmail.com> writes: what does this mean? how do I install 41.0.4 ? Firefox needs sqlite-3.8.11 which is currently in updates-testing but will go to stable in the next push. The bodhi link is https://bodhi.fedoraproject.org/updates/FEDORA-2015-16569 and you can download it from https://dl.fedoraproject.org/pub/fedora/linux/updates/testing/22/ (at least until it hits stable). In other words, some maintainer pushed a built with broken deps, Fedora's automated dep checkers and rel-eng also missed these. It's nice to see, how well Fedora's infrastructure works and how reliable it is ;) Maybe you should ask for a refund? No. I am simply expecting these guys to do their job, them not to push broken updates and them to have tools in place which prevent obvious cases like these. I am also expecting the people in charge to show up and take responsibility. I am also expecting them to analyse the cause to prevent such cases from happening again. Fedora is not in the situation 10 years ago. They do have tools in place (Or at least claim to have them) which should prevent cases like these. Oh, wait... :-) Ironically, I really do mean "wait". It will get sorted (and actually is now for me) and in the mean time you can use the version on your system that is still there. Nobody will die during this waiting period. If you check timestamps, you'll notice that - they pushed the sqlite update more than 12 hours after the firefox update. - they pushed the sqlite update several hours after people started to complain about the broken firefox update on this list. - The sqlite update did not go through the usual Fedora QA-process. In other words, as things appear, somebody was firefighting after the broken update had gone public. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: firefox?
On 09/27/2015 03:24 PM, Andre Robatino wrote: Paul Cartwright gmail.com> writes: what does this mean? how do I install 41.0.4 ? Firefox needs sqlite-3.8.11 which is currently in updates-testing but will go to stable in the next push. The bodhi link is https://bodhi.fedoraproject.org/updates/FEDORA-2015-16569 and you can download it from https://dl.fedoraproject.org/pub/fedora/linux/updates/testing/22/ (at least until it hits stable). In other words, some maintainer pushed a built with broken deps, Fedora's automated dep checkers and rel-eng also missed these. It's nice to see, how well Fedora's infrastructure works and how reliable it is ;) Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: OT: SSD or not to SSD, that is the question
On 09/05/2015 11:59 PM, Cameron Simpson wrote: Finally, I also recommend an SSD for your system drive. For /var, not so important, but the OS it is fantastic. Having /var (more precisely /var/lib/mock and /var/cache/mock) on SSD is important for mock. It speeds up mock-building rpms significantly. However, this price for this is pretty high, because mock adds quite significant amounts of additional diskspace requirements. Without mock, a Fedora installation easily fits on 32GB (and less), when working w/ mock you need much more. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: e2fsck after hard shutdown
On 09/05/2015 07:31 AM, antonio montagnani wrote: do you mean that I could run e2fsck from inside the root shell? Yes. I guessed that filesystems were already mounted and e2fsck doesn't work on mounted filesystems I have to study. :-) Depends, depends on what actually happened/gone wrong. In most such cases, the filesystems are mounted read-only or not mounted at all. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: OT: SSD or not to SSD, that is the question
On 09/04/2015 10:56 PM, Pete Travis wrote: On Sep 4, 2015 3:14 PM, "Richard Shaw" mailto:hobbes1...@gmail.com>> wrote: > > I have two drives in my desktop computer, a 500GB for / and /var which is failing, and a 1TB drive for /home and swap. > > Cost is definitely an issue so I have narrowed my options down to a 1TB drive for about $53 or a 256MB SSD for about $86. > > Obviously the SSD is about half the size of my current failing drive. I think I can squeeze things down a bit to make it fit and I REALLY want to get an SSD. I don't know, if this would be an option to you, but if I were you, I'd consider a smaller SSD in combination with a bigger HDD. Somewhat simplified, my partition scheme is /home on HDD and all the rest (including /var) on SDD. From my experience, using this scheme, a 128GB SSD is way more than sufficient for Fedora. A 64GB SSD would also work, but then diskspace on SSD will be tight - Certainly, YMMV ;) > With current reliability of MLC SSD's is it a good idea to put /var on an SSD? The major users there would be BackupPC and all the chroots/cache for mock packaging builds. I've been doing this for years with any problems (I don't use BackupPC, but am heavily using mock (on SDD)). Absolutely get a *quality* SSD. I've had a Samsung 830 since they were new and use mock and got and more on a frequent basis, with no ill effects. I share this experience. I have a Samsung 430, which still is in daily use today (32466hrs (3,7yrs) power-on-hours) and so far has not exposed any issues (fingers crossed :) ). Even if life expectancy were half that of a spinning drive instead of an order of magnitude more, I'd still get one for performance reasons. Fully agreed - I would not want to work on systems without SSD, anymore. They provide elderly HW with a performance boost and "liveliness", which may easily extend a system's life-time by years. If you had an SSD on this system before, I wouldn't expect you to be satisfied with a HDD-only system. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: PCI-e SSD recommendation for a F22 machine
On 09/04/2015 04:58 PM, Ranjan Maitra wrote: I used gcc with -O3 or -Ofast. Any suggestions on this? Stay with -O2 unless you precisely know what you are doing. Playing with -O3 and -Ofast may squeeze out some usecs somewhere, but because you are using less used and tested paths of the compiler, you're at risk of facing bugs/miscompilation etc. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: Sorting iPhone image files -
On 09/01/2015 03:47 PM, Markku Kolkka wrote: 1.9.2015, 16:31, Bob Goodwin kirjoitti: So it seems there probably is either software to sort them or a script that would sort on the date shown by "file." What can I do to accomplish this? Use exiftool to rename the files based on the creation date: http://www.sno.phy.queensu.ca/~phil/exiftool/#filename The corresponding Fedora-package is perl-Image-ExifTool. Try dnf install perl-Image-ExifTool or dnf install /usr/bin/exiftool Another helpful tool in cases like the OP's is jhead (Package jhead), a tool which allows to manipulate jpgs' exif data. I am using it to achieve similar purposes as the OP in shell scripts. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: More dnf annoyance
On 08/11/2015 04:53 PM, Matthew Miller wrote: On Tue, Aug 11, 2015 at 04:35:56PM +0200, Heinz Diehl wrote: Yet two completely separate contacts with Fedora's metalink server. Trouble-shooting these kinds of problems would need to include a closer look at what mirrors you are assigned to in both cases. Ok, I see. So what command should I use to keep my system updated? Usually, I update once a week (or two). Usually, just plain `dnf upgrade`. If there's an urgent update, you might want to worry about the kind of issues you're seeing here; otherwise, it'll eventually settle itself out. This was the case last weekend: firefox-39.0.3 Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: More dnf annoyance
On 08/11/2015 01:32 PM, Patrick O'Callaghan wrote: On Tue, 2015-08-11 at 13:13 +0200, Ralf Corsepius wrote: On 08/11/2015 12:51 PM, Patrick O'Callaghan wrote: On Tue, 2015-08-11 at 10:35 +0200, Heinz Diehl wrote: F22, in short: first running "dnf --refresh upgrade" shows some new packets. Then "dnf clean all" followed by "dnf --refresh upgrade" shows the same packets to be updated, and *some more*. So two update commands at different times give different results? IIUC, you are misunderstanding. The issues behind this are - "dnf --refetch" is refetching different versions of metadata from different (and differently sync'ed and/or broken) mirrors - fedora's mirrorlists are pointing to mirrors being out of sync. What matters is whether dnf is seeing the same state the two times it runs. In this case it clearly isn't. Exactly. With "dnf --refresh" it often does not see the same state. However, it should! The fact it does not see the same state, means https://mirrors.fedoraproject.org pushing bogus information. In addition to that, https://mirrors.fedoraproject.org seems to have been down and inaccessible for several hours, last weekend, which caused additional issues with dnf (and yum). Same result. If dnf is run twice you can't guarantee it will give the same result. If dnf and mirror-management was functional, then - except in those rare situations when the master has just been updated - they must point to mirrors carrying an identical state. That's an inherent feature of loosely distributed systems (where there isn't a distributed consensus protocol). Obviously the wider apart the two runs, the more differences will tend to appear, but the presence of differences does not in itself indicate a problem. My assumption is: mirror-manager is dysfunctional and dnf isn't sufficiently robust. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: More dnf annoyance
On 08/11/2015 12:51 PM, Patrick O'Callaghan wrote: On Tue, 2015-08-11 at 10:35 +0200, Heinz Diehl wrote: F22, in short: first running "dnf --refresh upgrade" shows some new packets. Then "dnf clean all" followed by "dnf --refresh upgrade" shows the same packets to be updated, and *some more*. So two update commands at different times give different results? IIUC, you are misunderstanding. The issues behind this are - "dnf --refetch" is refetching different versions of metadata from different (and differently sync'ed and/or broken) mirrors - fedora's mirrorlists are pointing to mirrors being out of sync. In addition to that, https://mirrors.fedoraproject.org seems to have been down and inaccessible for several hours, last weekend, which caused additional issues with dnf (and yum). Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: More dnf annoyance
On 08/11/2015 12:16 PM, Tom Horsley wrote: On Tue, 11 Aug 2015 10:35:04 +0200 Heinz Diehl wrote: F22, in short: first running "dnf --refresh upgrade" shows some new packets. Then "dnf clean all" followed by "dnf --refresh upgrade" shows the same packets to be updated, and *some more*. Last Sunday, I've had a case, where I resorted to rm -rf /var/cache/dnf because neither "dnf clean all" nor "dnf --refresh" seems to have worked. No matter what I did dnf seems have refetched the same outdated mirror presenting me the same updates. I don't think that's new with dnf. I've seen similar from yum. Well, IIRC, in its infancy yum has had similar issues. The common work around was to "yum clean metadata", then. Barring the fact Fedora mirrors seem to be broken quite often, these day, with dnf, the situation seems to have worsened. AFAICT, it doesn't correctly validate metadata and/or seems to prefer to refetch broken/outdated/dead mirrors. It all depends on which mirrors it picked to get the data from and the timing of mirrors getting updated. Which only means one thing - What I wrote above ;) Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: dnf update vs Software Udpates
On 07/23/2015 08:28 AM, Radek Holy wrote: - Original Message - From: "Ralf Corsepius" To: users@lists.fedoraproject.org Sent: Wednesday, July 22, 2015 6:58:49 PM Subject: Re: dnf update vs Software Udpates On 07/22/2015 05:41 PM, Heinz Diehl wrote: On 22.07.2015, Suvayu Ali wrote: I usually update weekly (or at least once within two weeks). And since F22, I get "nothing to do" every time I do this What you describe indicates you could be victim of what I conside a massive design flaw in dnf, the dnf guys have been ignoring ever since, because they believe to know better: When dnf encounters a broken dependency, it doesn't tell you about it and ignores it. Try "dnf --refresh --best update" Ralf Let's admit that you ignore us as well. No, I do not ignore you. I am simply tired of being confronted with this immature and broken piece of banana software called "dnf" and am tired of permanently being confronted with what I perceive as "false promises". Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: dnf update vs Software Udpates
On 07/23/2015 06:05 AM, Rahul Sundaram wrote: Hi On Wed, Jul 22, 2015 at 10:55 PM, dwoody5654 wrote: Is there a way to make dnf provide info instead of being silent? The answer was posted earlier in the thread. Well, the real answer would be to change dnf's behaviour. The current behaviour is just non-helpful design flaw. -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
Re: dnf update vs Software Udpates
On 07/22/2015 09:32 PM, Ron Yorston wrote: Matthew Miller wrote: On Wed, Jul 22, 2015 at 07:20:11PM +0100, Ron Yorston wrote: I certainly get the impression that dnf tells me about updates less frequently than yum did. It also seems to pull in metadata less frequently. Keep in mind that we only push updates once per day *anyway*. OK, but that's independent of the client being used. Both yum and dnf see the same updates, but dnf doesn't seem to be as quick to pass them on. IIRC, dnf and yum use different default intervals/timeouts. dnf by default uses 2 days (search for "metadata_expire" in "man dnf.conf", while yum used 6 hours (search for "metadata_expire" in "man yum.conf"). Of course, if you're mixing in private repos or other rpm providers, there may be different policies. Sure, different repos have different policies but why would that affect Fedora updates? They affect updates when package conflicts occur. The probability to encounter them when "mixing repos" is much higher. yum complained about them and forced users to intervene manually. dnf by default remains silent about such conflicts and doesn't update. This lets users believe everything was "OK", while their system actually is partially outdated. Ralf -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org