Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
The Politics of systemd
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3 ... , 28, 29, 30  Next  
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Gentoo Chat
View previous topic :: View next topic  
Author Message
Yamakuzure
Advocate
Advocate


Joined: 21 Jun 2006
Posts: 2170
Location: Bardowick, Germany

PostPosted: Wed Oct 14, 2015 9:05 am    Post subject: Reply with quote

gwr wrote:
Does nofail even make sense?
Yes. You can add it to ZRAM swap devices for instance. When fstab is processed, those devices might not exist, but get added later by sys-block/zram-init.

And prior the current versions, you had to add your zfs mount points to the fstab (not necessary any more) which would fail the moment fstab got processed, but were necessary later for mounting the remaining drives from your zpool(s).

Then you might have NFS shares that might appear later, but need to have an entry in fstab. However, I am not sure this is still needed, its been years since I used those.
_________________
elogind
CygwinOnSteroids
"A conservative is a man who is too cowardly to fight and too fat to run."
-- Elbert Hubbard
Back to top
View user's profile Send private message
gwr
Apprentice
Apprentice


Joined: 19 Nov 2014
Posts: 194

PostPosted: Wed Oct 14, 2015 9:16 am    Post subject: Reply with quote

Yamakuzure wrote:
gwr wrote:
Does nofail even make sense?
Yes. You can add it to ZRAM swap devices for instance. When fstab is processed, those devices might not exist, but get added later by sys-block/zram-init.

And prior the current versions, you had to add your zfs mount points to the fstab (not necessary any more) which would fail the moment fstab got processed, but were necessary later for mounting the remaining drives from your zpool(s).

Then you might have NFS shares that might appear later, but need to have an entry in fstab. However, I am not sure this is still needed, its been years since I used those.


Why would I use nofail when I could just remove the entry from fstab?
Back to top
View user's profile Send private message
Goverp
Guru
Guru


Joined: 07 Mar 2007
Posts: 561

PostPosted: Wed Oct 14, 2015 9:38 am    Post subject: Reply with quote

Anon-E-moose wrote:
Goverp wrote:
So why does that elect Hubbs to such an exalted minority?


My view of wh's incompetence isn't based on what he's done with openrc this time, but his general history with gentoo software

So it's a gratuitous insult. Not very nice.
_________________
Greybeard
Back to top
View user's profile Send private message
arnvidr
Guru
Guru


Joined: 19 Aug 2004
Posts: 574
Location: Oslo, Norway

PostPosted: Wed Oct 14, 2015 9:52 am    Post subject: Reply with quote

gwr wrote:
Yamakuzure wrote:
gwr wrote:
Does nofail even make sense?
Yes. You can add it to ZRAM swap devices for instance. When fstab is processed, those devices might not exist, but get added later by sys-block/zram-init.

And prior the current versions, you had to add your zfs mount points to the fstab (not necessary any more) which would fail the moment fstab got processed, but were necessary later for mounting the remaining drives from your zpool(s).

Then you might have NFS shares that might appear later, but need to have an entry in fstab. However, I am not sure this is still needed, its been years since I used those.


Why would I use nofail when I could just remove the entry from fstab?


I'd want it to continue trying to boot regardless. In my mind "nofail" implies "this cannot fail", which seems all kinds of backwards, but as I understand it, this flag is an old one, perhaps badly named? No matter, what I'm getting is, is that it would make heaps more sense to mark any special drives that you would want to halt the boot process if they failed to mount. As already mentioned, any really critical drives would bork the boot anyway, so if you wanted anything non-critical to error out the boot, put a mark on them.
_________________
Noone wrote:
anything
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5083
Location: The Peanut Gallery

PostPosted: Wed Oct 14, 2015 10:09 am    Post subject: Reply with quote

digi_owl wrote:
Nofail just suppress the errorcode on a failed mount command.

WTF? How on Earth is that going to help?

Talk about lunacy.
Quote:
Thing is that previous inits have basically ignored mount error codes unless explicitly told otherwise.

So let's start ignoring them at fstab level? Lul.

Only in systemdbust-land would you get "turn off errors in the fstab" as a solution to "ignoring mount error codes".
Quote:
Thing is that / is mounted earlier than most, so of it fails you will not even get into the general mount step of the init sequence.

How is that "the thing"? We've mentioned this above: "We're only talking about the occasional times where there's an error, not the usual case for partitions on the same drive as your rootfs and /boot, and hopefully not for more esoteric setups either (or they need to sort out the bigger problems, first.)"

Note: I am not saying "never use nofail"; it has its uses as Yamakuzure showed.
Quote:
While /bin and /sbin are directories on /, as long as / mounts the rest can be recovered from within the live system.

But now there is a push to move /(s)bin into /usr, perhaps even punt /usr onto NAS or SAN.

Just because there is a putsch toward only supporting the same use-case as upstream systemdbust this year, does not mean we should all blindly follow.
People have been using /usr-on-NFS et al. for an awfully long time before these numbskulls turned up..
Quote:
current low level user space developers only see two kinds of computers. Desktops/laptops and container/VM farms.

They're not "low-level developers", though; they're desktop developers who think they're low-level because they don't just code in javascript.

As someone else pointed out in a debian thread, many of them don't even run Linux; they use Mac-OS and virtualisation, which is why they're so shoddy: they don't care about critical bugs, like their users do, as they simply reboot the VM.
Which also explains why they bang on about the "cloud" and reboot times for VMs: that's what they do a lot of, unlike admins on real machines.
Quote:
.. Anything else do not enter into the equation.

Which is entirely the problem with their blinkered "vision".
Quote:
Then again it all seems to come down to being able to take automated action upon the appearance of a new storage device in /dev. to me at least that sounds like autorun on steroids...

As usual, a use-case no-one was talking about before, since this was all aimed at the poor luser on their desktop, who can only handle a GUI, and only if you dumb it down supposedly, remember?
As pointed out above, VM farms etc should be treated as a different use-case, in the same way thin-clients were handled in the round by the development of LTSP, and should not be done via a half-assed solushion like loginkit.

As you said, better to make use of existing solutions like autorun, which do not try to be more than they are, or should be.

Better still to flip the bozo bit on these bozos, and get on with fun stuff that doesn't rot your brain and your CPU.
Back to top
View user's profile Send private message
Yamakuzure
Advocate
Advocate


Joined: 21 Jun 2006
Posts: 2170
Location: Bardowick, Germany

PostPosted: Wed Oct 14, 2015 12:24 pm    Post subject: Reply with quote

gwr wrote:
Yamakuzure wrote:
gwr wrote:
Does nofail even make sense?
Yes. You can add it to ZRAM swap devices for instance. When fstab is processed, those devices might not exist, but get added later by sys-block/zram-init.

And prior the current versions, you had to add your zfs mount points to the fstab (not necessary any more) which would fail the moment fstab got processed, but were necessary later for mounting the remaining drives from your zpool(s).

Then you might have NFS shares that might appear later, but need to have an entry in fstab. However, I am not sure this is still needed, its been years since I used those.


Why would I use nofail when I could just remove the entry from fstab?
In case were you *MUST* have an entry in fstab. When you do not need it (like with zfs these days) there is absolutely no reason to add an entry with 'nofail' to the fstab.

And what about network shares that might not be online, yet? But you want them to be automatically mounted?

"nofail" does not mean that the error is suppressed and the mounting fails silently. The device *is* mounted then, no matter if available or not. This is, of course, only good for devices which show up later.
_________________
elogind
CygwinOnSteroids
"A conservative is a man who is too cowardly to fight and too fat to run."
-- Elbert Hubbard
Back to top
View user's profile Send private message
arnvidr
Guru
Guru


Joined: 19 Aug 2004
Posts: 574
Location: Oslo, Norway

PostPosted: Thu Oct 15, 2015 8:50 am    Post subject: Reply with quote

Yamakuzure wrote:
"nofail" does not mean that the error is suppressed and the mounting fails silently. The device *is* mounted then, no matter if available or not. This is, of course, only good for devices which show up later.
Ah, so mounting is completely broken with openrc 0.18 then.
_________________
Noone wrote:
anything
Back to top
View user's profile Send private message
Yamakuzure
Advocate
Advocate


Joined: 21 Jun 2006
Posts: 2170
Location: Bardowick, Germany

PostPosted: Thu Oct 15, 2015 10:45 am    Post subject: Reply with quote

arnvidr wrote:
Yamakuzure wrote:
"nofail" does not mean that the error is suppressed and the mounting fails silently. The device *is* mounted then, no matter if available or not. This is, of course, only good for devices which show up later.
Ah, so mounting is completely broken with openrc 0.18 then.
openrc has nothing to do with it.

It's like certain kernel options. If you do not know what it's good for, you do not need it. Otherwise you go the Poettering way: "I dunno what it is, so it's broken!"... :lol:
_________________
elogind
CygwinOnSteroids
"A conservative is a man who is too cowardly to fight and too fat to run."
-- Elbert Hubbard
Back to top
View user's profile Send private message
Anon-E-moose
Advocate
Advocate


Joined: 23 May 2008
Posts: 2854
Location: Dallas area

PostPosted: Thu Oct 15, 2015 10:49 am    Post subject: Reply with quote

Maybe this is WH's way of "destroying" openrc so that the only alternative is sysd.
Yes, I believe he's just that despicable.

Fortunately we (the users) can fork openrc or use whatever version we want, the hell with what he wants.
_________________
Asus m5a99fx, FX 8320 - amd64-no-multilib, 4.14.52-zen, glibc-2.26-r7, gcc-7.3.0, eudev, openrc-0.13.11
xorg-server-1.19.5-r1 w/mesa-17.3.9, openbox, nouveau and radeon, oss4(2017), profile 17.0 (no-pie)
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5083
Location: The Peanut Gallery

PostPosted: Mon Oct 19, 2015 5:55 pm    Post subject: Reply with quote

steveL wrote:
Council and Comrel pretend not to see the conflict-of-interest that is blatantly obvious to everyone else, in having the developers oversee all social issues, despite it being completely outside the Council's mandate and area of expertise.

As a result, developers continue to abuse the process, filing complaints against anyone they don't like, which never result in any comeback when they fail miserably. ("Better luck next time," seems about the totality of it.)

A Council member like William Hubbs, can and has abused the same process, and shows up trying to push the same agenda without mentioning that it's a personal issue which he wants to pursue. This raises the question of what you're supposed to do when a Council member is way out of control, to which the answer would appear to be: capitulate.

Despite being told over and over that he misinterpreted what was said about 2 years ago, Hubbs continues to abuse his position, trying to get the Trustees or the Council to agree to remove me, without even a semblance of due process: just on his say-so (because he's on the Council, presumably.)
For instance, he had antarus (Alec Warner) raise it with Trustees, who didn't see what it had to do with them; without any notification to me, of course.
If it had gone through as he'd hoped (and he was clearly priming antarus) I would have been summarily removed from the Foundation, without even knowing it had occurred.

Then he brought it up on the project mailing-list, to put on the Council agenda; without mentioning the real background ofc, even when it was queried.
...
Feel free to find the discussions referred to.. if you think I'm lying.

NeddySeagoon wrote:
Foundation meeting logs are linked from https://wiki.gentoo.org/wiki/Foundation:Main_Page#Principles_of_the_Gentoo_Foundation.
See under Board Minutes.

Cheers, Neddy.

Here's the Foundation meeting referred to.

I'm not going to dignify the offensive characterisations with discussion; every time I read it, it makes me nauseated at the levels which apparatchiks will descend to.

And make no mistake: that's exactly what is happening here. Using a specious basis of "the Developer CoC has a Community mandate to apply to everyone", which as discussed previously is a complete rewriting of history (ie: false), the apparatchiks are trying to do an inverse takeover (leverage, remember?) via the frankly ultra vires[1] device of claiming the Council has final say over social matters; despite it only ever being setup to handle technical matters, which has always been the "pure" basis for the Council to be elected in the first place.

This assumption can be seen in-play, as the assumed background to Hubbs' bringing up CoC violations applying to Foundation members, which as we can see from the Foundation log above, is simply a continuation of his underhand attack on me.

Just bear in mind that the "alleged offence" occurred well over a year before that meeting, during which time I have had zero interaction with Hubbs. So even if the offence had not been consistently denied, and had in fact been found, this would not have constituted a "strike" under either CoC (the Community one, or the subsequent reduced Developer one): that requires an ongoing pattern of behaviour, and refusal to accept moderation, which was never even attempted, over a period of months.
Instead a move was made to ban forthwith, label it a strike, and refer to an incident from over 5 years ago, which had never even been called a strike by anyone involved, falsely as such, and thus to threaten that I was on 2 strikes; ie: if we can catch you out once more, we'll ban you permanently from all Gentoo media, despite the clear abuses of process.

As we've seen with debian, a "technical committee" makes a perfect attack vector for corporations seeking to coopt community work and sell it back to the people who developed it: its users.
Precisely because they pride themselves on being "purely technical" while having no clue about social matters, "technical committees" are incredibly easy to manipulate with social pressure; aka propaganda, aka advertising and marketing. Even more so once you have coopted or placed a couple of its members.

Again, it's up to the Foundation to ensure that the purely theoretical (when it comes to legal effect) Council, does not act ultra vires, as anything else would be a clear denial of its own legal responsibilities. You cannot delegate authority over X to an internal group, constituted solely to deal with X, and then allow them to mess up Y, especially when Y is central to your mission (in this case: Y is the community and social relations across it.)
"Mission-creep" (the usual euphemism for acting outside one's remit) is clearly the method of choice for people seeking to undermine the pillars of Gentoo, and it is up to the Foundation to ensure the pillars are upheld; or the notion of an Officer has no meaning, since its legal basis is void.
Quote:
I'll look into the rewording of the Foundation Charter too.

Is there a vcs commit history to show who changed the document from a Charter to "something else" (undefined)?

==
[1] ultra vires simply means "acting outside a given mandate or domain", but it means the actions carried out are legally void, as well as other consequents, usually applied to whomever has carried out the actions, and always to the organisation concerned: especially if done knowingly, which this clearly is since the matter has been raised on so many occasions.
The usual weasel response to this is "the Council doesn't even exist legally" (which is precisely why no-one argues with the assertion as made above); as we've seen however, this clearly can lead to a failure by the Foundation in maintaining its own Charter, which is what gives every other activity any legal basis.
It certainly has proven so in the case of Debian, where a technical committee has ended up making decisions outside its scope, with ramifications that impinge on the first fundamental pillar of Gentoo: user choice.

Hubbs is certainly of the new-school in acting blasé about removing user-choice; as we've seen, simply so he can follow trends in what can only be described as an idiotic upstream.

(The latter reads like a post-facto rationalisation for not even notifying the Council of the option to run udev with a split /usr and no initramfs, via simple initscript patches, ie not modifying the precious "upstream" code.)
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5083
Location: The Peanut Gallery

PostPosted: Mon Oct 19, 2015 6:16 pm    Post subject: Reply with quote

gwr wrote:
Quote:
[17:16:34] <igli> worrying part is adding redirection like: '< /dev/null > /dev/null' to command_args

Well, if they're lying to their scripts about input and output, then no wonder they have so much trouble with shell scripting. If a script is _disallowed_ to process input or output, better to use -1 and -2 so that it receives an error. It would be worse to port a script over and fool it into thinking it is not getting input or is doing output. Who knows what that script is doing?

I was expecting a rejoinder along the lines of "it's ok, we pass those through eval", from where I wanted to move to the kind of discussion you're having.
The actual issue isn't such a big deal; it's more the mismatch when you realise you're talking to someone who doesn't even know the basics. That's fine somewhere like #bash or #awk from someone seeking to learn, but not from someone who's supposedly the lead of a shell and C project; certainly not after over 3 years on the job.

Let alone someone who acts so condescendingly in ignoring users till they go away, but falls over himself to ensure other developers feel a warm sense of obligation towards him. The former behaviour is what led me to "rage-quit", in the end: I just couldn't stomach it any more.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5083
Location: The Peanut Gallery

PostPosted: Mon Oct 19, 2015 6:22 pm    Post subject: Reply with quote

steveL wrote:
It's absolutely a generic "you"

Tony0945 wrote:
I'm relieved to hear it. Sometimes this is not clear and I think it gets you into trouble. You are a man of passion and I understand that for I am a passionate man also.

Well, thanks for picking me up on the ambiguity; it only improves the discourse.

I wanted to think about your last comment for a bit before responding.

It seems to me that sociopaths (for want of a better word) take pleasure in provoking an emotive reaction from other human-beings, then using that reaction as a justification for action if possible, and deflection if not.

It also appears to me that most apparatchiks are sociopaths; not all, by any means: I've met some organisation geeks who aren't at all sociopathic in any form. Just that most people who dedicate their lives to climbing the tree and nothing else, are in fact missing some vital human connection.

I think that's why they suddenly cave, in so many situations: they were just trying a play, and when it fails, they have the next one lined up, which they move on to with aplomb, taking the opportunity to appear "cool" in the process.
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 3340

PostPosted: Thu Oct 22, 2015 5:57 pm    Post subject: Reply with quote

I'm a bit surprised to get a few days down the road, and nobody has posted a link to this:

http://phoronix.com/scan.php?page=news_item&px=Systemd-Architecture-Read

which leads to this:

http://blog.darknedgy.net/technology/2015/10/11/0/

Obviously the top article is Phoronix, a pro-sysemd site. But the referenced link is quite interesting. It's a technical dive into systemd, and not complimentary at all. But because it skipped any Unix philosophy arguments, it seems to get a reading from Phoronix, and not instant vitriol. It also seems to be long and deep enough that it flummoxed a lot of Phoronix readers, diverting complaints.

On the more interesting side, it recognizes that systemd is far more than an init system, and takes some time looking at how well (or how poorly) it does at biting off all of that.

Incidentally, it's by the guy who was developing uselessd, and eventually dropped it when he realized that the real scope of systemd was far beyond an init system, and couldn't readily be scaled back.
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
arnvidr
Guru
Guru


Joined: 19 Aug 2004
Posts: 574
Location: Oslo, Norway

PostPosted: Fri Oct 23, 2015 7:54 am    Post subject: Reply with quote

depontius wrote:
Obviously the top article is Phoronix, a pro-sysemd site.
Really? Isn't this the site that was so negative to systemd a year or so back that it got some serious backlash from some of its readers, and it now reports on systemd in a more or less 'sarcastic neutral' way?
_________________
Noone wrote:
anything
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 3340

PostPosted: Fri Oct 23, 2015 10:38 am    Post subject: Reply with quote

arnvidr wrote:
depontius wrote:
Obviously the top article is Phoronix, a pro-sysemd site.
Really? Isn't this the site that was so negative to systemd a year or so back that it got some serious backlash from some of its readers, and it now reports on systemd in a more or less 'sarcastic neutral' way?


I hadn't thought of it that way, perhaps you're right. However the general comment tone tends to be very pro-systemd. Surface opposition, particularly Unix philosophy, and rapidly get crisped, at least textually.

Still, the original article was very insteresting, and not pro-systemd at all. But it stayed in terms that either the pro-systemd commenters couldn't oppose or couldn't understand.
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
Yamakuzure
Advocate
Advocate


Joined: 21 Jun 2006
Posts: 2170
Location: Bardowick, Germany

PostPosted: Fri Oct 23, 2015 11:26 am    Post subject: Reply with quote

depontius wrote:
http://blog.darknedgy.net/technology/2015/10/11/0/
That's a great article!

My favourite parts so far:
Quote:
There is at least one executable unit which is solely transient, that of scopes – which are a form of logical binding to associate multiple non-forking tasks in a cgroup for resource management purposes, not unlike how svchost does grouping, but with different mechanics.
svchost ... So this will give people fuel and a new spring to their step who always claimed RedHat wants to build another windows. :D
Quote:
Mount units, in fact, serve a similar purpose to a job or oneshot service, and even try to resemble jobs in emitting job status messages, but are internally implemented on top of the generic Unit interfaces as what boils down to a driver for the mount binary in util-linux. The same is true of swap units for swapon/swapoff, but along the way also use libudev interfaces to register a device node and actually do seem to queue jobs. This makes these two types somewhat of a duplication and overlapping case, on top of hiding inconsistency.
Yeah. They are good in wrapping inconsistencies around long existing tools without adding value, eh? ;)
Quote:
For one thing, none of jobs, transactions, unit semantics or systemd-style dependencies map to the Unix process model, but rather are necessary complications to address issues in systemd being structured as an encapsulating object system for resources and processes (as opposed to a more well-defined process supervisor) and one accommodating for massive parallelism.
So basically the systemd architecture and design is useless to achieve systemd's own goals? Nice! Or in other words: A lot of resources have been spent to solve problems that wouldn't have been there without systemd.
Quote:
Stuck jobs
Where single or multiple start jobs take excessively long times (30s, >1m, etc.) to synchronously complete (likely due to being an anchor) before moving to the next. Generally resolved by shotgun debugging via disabling the culprit, as the information in systemd-analyze(1) is not granular enough to make estimations about the internal scheduling behavior.
So they wrote a system they can not debug but have to shoot down parts of it (with a shrug I imagine) because they were not able to write analytics about their own software? Wow.
Quote:
The ultimate irony then becomes that, one must communicate with an object system (D-Bus) in order to launch commands or query data which must be deserialized to another object system, all the while the real object system is obscured, non-uniform and hidden from user manipulation, though the proxy object system (D-Bus) nonetheless increasing overhead and failure paths of communication.
Hear, hear!
Quote:
In fact, for an object system, there is indeed a general lack of delegation or being able to rebind and extend options. In fact, systemd, unlike prior systems such as pinit, initng and eINIT, plus later ones like finit, does not have anything resembling a plugin or extension system despite its high surface and potential for needing to customize its options.
Which looks like supporting the worries that the systemd devs try to shut the users out.
Quote:
The object system is mostly a mock, and amounts to little more than a smoke screen, only complicating the execution model through excessive indirection without tangible benefit.
I think this is a nice aggregation of quite a few paragraphs and arguments written by SteveL, isn't it? You were always right, man! :lol:

Really, reading that article was totally worth the time. :)

Thank you again, depontius!
_________________
elogind
CygwinOnSteroids
"A conservative is a man who is too cowardly to fight and too fat to run."
-- Elbert Hubbard
Back to top
View user's profile Send private message
Anon-E-moose
Advocate
Advocate


Joined: 23 May 2008
Posts: 2854
Location: Dallas area

PostPosted: Fri Oct 23, 2015 3:58 pm    Post subject: Reply with quote

Interesting :wink:
_________________
Asus m5a99fx, FX 8320 - amd64-no-multilib, 4.14.52-zen, glibc-2.26-r7, gcc-7.3.0, eudev, openrc-0.13.11
xorg-server-1.19.5-r1 w/mesa-17.3.9, openbox, nouveau and radeon, oss4(2017), profile 17.0 (no-pie)
Back to top
View user's profile Send private message
Shamus397
Apprentice
Apprentice


Joined: 03 Apr 2005
Posts: 215
Location: Ur-th

PostPosted: Fri Oct 23, 2015 5:09 pm    Post subject: Reply with quote

To me, he summed up the whole sad state of affairs with what seems on the surface to be a throw-away line:
Quote:
Circular designs also hinder composability due to being all-or-nothing.

:D
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 3340

PostPosted: Fri Oct 23, 2015 5:39 pm    Post subject: Reply with quote

Conspiracy theory time. This really is a tin-hatter, because personally I think that systemd has been done by people originally trained on Windows who are uncomfortable with Linux, and want to "make it right", according to their definition of right.

However, just because there's not conspiracy, just because it may be the innocent action of (we consider them deluded) people, that's not to say the results might not look like a conspiracy.

So this article today pointed to by Ars Technica: http://techrights.org/2015/10/22/all-things-open-hijacked/

Microsoft up to its old games, again. But what if systemd happens to infringe on a Microsoft patent? Obviously it's in Microsoft's best interest to see systemd take over the Linux ecosystem, for alternatives to wither - no way back. Then sue for patent infringement.
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
Tony0945
Advocate
Advocate


Joined: 25 Jul 2006
Posts: 2496
Location: Illinois, USA

PostPosted: Fri Oct 23, 2015 6:18 pm    Post subject: Reply with quote

gwr wrote:
digi_owl wrote:
Nofail just suppress the errorcode on a failed mount command.

Thing is that previous inits have basically ignored mount error codes unless explicitly told otherwise.


This seems like the saner behaviour to me.


Oh yes! I'm so glad I read this thread. On several systems there are hard drives that I want to mount if I can, but don't want to stop the boot. One example is a dual boot system on which I want to automount the Windows hard disk if it can, but NOT let a failure stop booting Linux. On two others there are large disks for storage of videos. Again, I want those available, but do NOT want to halt booting if they can't be mounted.

I followed the instructions in the news bulletin adding "nofail" to the fstab but I certainly want notification if there is a problem. "noauto" is a rather crippling alternative. Do i have to drop into a shell and mount the drive every time I want to watch a video?

"non-critical" would be a better flag, but if I can't mount /boot then i would never have gotten to that point, likewise with / or /usr if I had a seperate /usr. So what's the point of stopping the boot for non-critical mounts?

Yes, I'm going to fix my /etc/fstab (i.e. remove nofail) and roll back OpenRC. And yes, that seems to be a deliberate crippling of OpenRC.


Last edited by Tony0945 on Fri Oct 23, 2015 11:49 pm; edited 1 time in total
Back to top
View user's profile Send private message
Roman_Gruber
Advocate
Advocate


Joined: 03 Oct 2006
Posts: 3806
Location: Austro Bavaria

PostPosted: Fri Oct 23, 2015 8:14 pm    Post subject: Reply with quote

It seems those systemd related support topics here have nearly vanished.

Is this now useable or have volks already switched to something else?

I may will download a bootable iso with systemd around xmas to have an insight. Some nubutu or some other live iso will have systemd anyway probably.
Back to top
View user's profile Send private message
Tony0945
Advocate
Advocate


Joined: 25 Jul 2006
Posts: 2496
Location: Illinois, USA

PostPosted: Fri Oct 23, 2015 11:48 pm    Post subject: Reply with quote

tw04l124 wrote:


I may will download a bootable iso with systemd around xmas to have an insight. Some nubutu or some other live iso will have systemd anyway probably.
I installed Ubuntu to build a package that I had trouble with on Gentoo. The author told me that it built on Ubuntu. The Ubuntu installation was very fast, possibly because I had already prepared the partition with gparted from Gentoo. And the program did build on Ubuntu. However, Ubuntu took FOREVER to boot after selecting the chainload on the Gentoo menu. I attribute that to systemd/gnome 3. The gnome3 was unusable. I couldn't figure out how to get an xterm so I CTRL-ALT-F1's to a real terminal and built it there.
Back to top
View user's profile Send private message
arnvidr
Guru
Guru


Joined: 19 Aug 2004
Posts: 574
Location: Oslo, Norway

PostPosted: Sat Oct 24, 2015 7:07 am    Post subject: Reply with quote

depontius wrote:
arnvidr wrote:
depontius wrote:
Obviously the top article is Phoronix, a pro-sysemd site.
Really? Isn't this the site that was so negative to systemd a year or so back that it got some serious backlash from some of its readers, and it now reports on systemd in a more or less 'sarcastic neutral' way?


I hadn't thought of it that way, perhaps you're right. However the general comment tone tends to be very pro-systemd. Surface opposition, particularly Unix philosophy, and rapidly get crisped, at least textually.
Yeah, I think you're right on that, the commenters have always had a lot of pry-systemd people, but the main guy who runs the site has no love for it at all, unless he's recently changed his opinion of it.
_________________
Noone wrote:
anything
Back to top
View user's profile Send private message
GFCCAE6xF
Apprentice
Apprentice


Joined: 06 Aug 2012
Posts: 265

PostPosted: Sat Oct 24, 2015 8:32 am    Post subject: Reply with quote

arnvidr wrote:
depontius wrote:
arnvidr wrote:
depontius wrote:
Obviously the top article is Phoronix, a pro-sysemd site.
Really? Isn't this the site that was so negative to systemd a year or so back that it got some serious backlash from some of its readers, and it now reports on systemd in a more or less 'sarcastic neutral' way?


I hadn't thought of it that way, perhaps you're right. However the general comment tone tends to be very pro-systemd. Surface opposition, particularly Unix philosophy, and rapidly get crisped, at least textually.
Yeah, I think you're right on that, the commenters have always had a lot of pry-systemd people, but the main guy who runs the site has no love for it at all, unless he's recently changed his opinion of it.


He just tailors his titles and articles to get as many clicks as possible, his "views/opinions" also change about nvidia, amd, gnome, KDE, ubuntu, fedora, wayland, mir, how many BSD benchmarks to do, etc etc.
Hating systemd isn't the hottest meme to get ad money this year like it was during the Debian debate so writing articles about it and every little thing doesn't pull in attention to make it worth it.
Back to top
View user's profile Send private message
pdkl95
n00b
n00b


Joined: 14 Sep 2009
Posts: 4

PostPosted: Sat Oct 24, 2015 11:57 am    Post subject: Reply with quote

depontius wrote:
... personally I think that systemd has been done by people originally trained on Windows ... Microsoft up to its old games, again. ...


(As a long time gentoo user and forum lurker, hi everybody! Thank you for keeping the UNIX flame burning when so many others were distracted by shiny baubles!)

Re: Microsoft and other threats against Free Software. I just wanted to mention the old wisdom that protecting against security threats tends to also protect against mundane bugs. It doesn't necessarily matter if a security threat actually exists - the bugs should still be fixed.

In the unlikely case anyone hasn't seen it yet, I encourage watching PHK's Operation Orchestra, which discusses exactly these kinds of threats to Free Software. It doesn't matter if Microsoft or the NSA or Red Hat are actually involved; protecting against organized threats also protects against "people originally trained on Windows".
Back to top
View user's profile Send private message
Display posts from previous:   
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Gentoo Chat All times are GMT
Goto page Previous  1, 2, 3 ... , 28, 29, 30  Next
Page 29 of 30

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum