[opensuse-factory] SysV service error during dup on TW
Today when I ran 'zypper dup' the following error appeared: ( 77/307) Installing: man-2.7.6-6.1.x86_64 ...........[done] Additional rpm output: SysV service man-db-create does not exist, skipping Updating /etc/sysconfig/cron ... Since TW has been using 'systemd' for quite a while why would the man package still be trying to run a 'SysV' script? -- Ken Schneider -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
20.12.2017 19:47, Ken Schneider - Factory пишет:
Today when I ran 'zypper dup' the following error appeared:
( 77/307) Installing: man-2.7.6-6.1.x86_64 ...........[done] Additional rpm output: SysV service man-db-create does not exist, skipping Updating /etc/sysconfig/cron ...
Since TW has been using 'systemd' for quite a while why would the man package still be trying to run a 'SysV' script?
What makes you think it is "trying to run a 'SysV' script"? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, Dec 20, 2017 at 1:47 PM, Ken Schneider - Factory <suse-list3@bout-tyme.net> wrote: why would the man
package still be trying to run a 'SysV' script?
It isn't. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 12/20/2017 08:54 PM, Cristian Rodríguez wrote:
On Wed, Dec 20, 2017 at 1:47 PM, Ken Schneider - Factory <suse-list3@bout-tyme.net> wrote: why would the man
package still be trying to run a 'SysV' script?
It isn't.
Then what created this output: SysV service man-db-create does not exist, Looks like it clearly says something about a 'SysV service' -- Ken Schneider SuSe since Version 5.2, June 1998 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi Ken! On 12/21/2017 07:09 PM, Ken Schneider - openSUSE wrote:
Then what created this output:
SysV service man-db-create does not exist,
Looks like it clearly says something about a 'SysV service'
I think you're misinterpreting this message.
From your original post:
( 77/307) Installing: man-2.7.6-6.1.x86_64 ...........[done] Additional rpm output: SysV service man-db-create does not exist, skipping Updating /etc/sysconfig/cron ... This means that the postinst script from the "man" package says that your system doesn't have the SysV service "man-db-create" which is why it's not trying to run the script in the first place. This is just an information which you can ignore. The "man-db-create" SysV script doesn't exist on your machine because your system doesn't have SysVInit in the first place. It will just run the systemd equivalent command to create the database for manpages. So, you can just ignore this message. Adrian -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hello, an addedum only FYI: On Dec 21 19:24 John Paul Adrian Glaubitz wrote (excerpt):
On 12/21/2017 07:09 PM, Ken Schneider - openSUSE wrote:
Then what created this output:
SysV service man-db-create does not exist,
Looks like it clearly says something about a 'SysV service'
I think you're misinterpreting this message.
From your original post:
( 77/307) Installing: man-2.7.6-6.1.x86_64 ...........[done] Additional rpm output: SysV service man-db-create does not exist, skipping Updating /etc/sysconfig/cron ...
This means that the postinst script from the "man" package says that your system doesn't have the SysV service "man-db-create" which is why it's not trying to run the script in the first place.
This is just an information which you can ignore. The "man-db-create" SysV script doesn't exist on your machine because your system doesn't have SysVInit in the first place. It will just run the systemd equivalent command to create the database for manpages.
So, you can just ignore this message.
I think messages that are meant to be ignored by the user (i.e. messages that are meant only as optional information) should not be shown to the user by default. Normally such messages may appear only in a log file or are shown to the user only in debug or verbose mode. I don't know how excatly Ken Schneider ran 'zypper dup' i.e. with or without '--verbose' or '--details'. Cf. "WARNING is a waste of my time" at http://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html You may also have a look at its origin https://github.com/rear/rear/issues/564 therein in particular starting at https://github.com/rear/rear/issues/564#issuecomment-86188528 and subsequent comments. Kind Regards Johannes Meixner -- SUSE LINUX GmbH - GF: Felix Imendoerffer, Jane Smithard, Graham Norton - HRB 21284 (AG Nuernberg) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 12/22/2017 09:33 AM, Johannes Meixner wrote:
I think messages that are meant to be ignored by the user (i.e. messages that are meant only as optional information) should not be shown to the user by default.
Normally such messages may appear only in a log file or are shown to the user only in debug or verbose mode.
I don't know how excatly Ken Schneider ran 'zypper dup' i.e. with or without '--verbose' or '--details'.
Those messages usually show up as warning because it might not always be known from the programmer's point of view whether the reported issues might be a problem or not. Warning basically says "I think this particular issue could cause an issue on your system but that's not 100% sure so I'll just inform you but let you continue anyway".
Cf. "WARNING is a waste of my time" at http://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html
Same here. Take gcc, for example. The compiler will warn you in many cases when you forget to initialize variables or perform a cast with incompatible data types which could result in your code crashing. However, since neither are a violation of the C language specification, the compiler will just let you continue. Sometimes the lack of initialization or the casts are intended by the programmer. And in this case, Ken was also misreading the message. It clearly said there is no SysV script on the system, yet he assumed that was the case. Plus, the warning is actually a hint to the programmer that this particular piece of SysV code should be removed. Adrian -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hello John Paul Adrian Glaubitz, I think you may have missed that I was talking about what to show to the user _by_default_ and the reasoning behind http://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html On Dec 22 10:00 John Paul Adrian Glaubitz wrote (excerpt):
Those messages usually show up as warning because it might not always be known from the programmer's point of view whether the reported issues might be a problem or not.
Of course, but see https://github.com/rear/rear/issues/564#issuecomment-86418677
Warning basically says "I think this particular issue could cause an issue on your system but that's not 100% sure so I'll just inform you but let you continue anyway".
Of course, but see https://github.com/rear/rear/issues/564#issuecomment-86458179 Kind Regards Johannes Meixner -- SUSE LINUX GmbH - GF: Felix Imendoerffer, Jane Smithard, Graham Norton - HRB 21284 (AG Nuernberg) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 12/22/2017 01:26 PM, Johannes Meixner wrote:
Those messages usually show up as warning because it might not always be known from the programmer's point of view whether the reported issues might be a problem or not.
Of course, but see https://github.com/rear/rear/issues/564#issuecomment-86418677
Warning basically says "I think this particular issue could cause an issue on your system but that's not 100% sure so I'll just inform you but let you continue anyway".
Of course, but see https://github.com/rear/rear/issues/564#issuecomment-86458179
This isn't about shifting responsibility, this is simply a result of the fact that most software simply cannot cover all potential configurations and use cases that exist out there. I don't understand why some people have to question fundamental design principles of software just because one piece of software doesn't cover their particular use case. "I personally found warning messages to be pointless for my use cases, so they must be useless for everyone and every piece of software on the planet." No one is hurt by a warning, but in some cases it gives you a hint what to look for if you run into weird issues. You always have to keep in mind that there are millions of different use cases and configurations out there, including those people who use different init systems like OpenRC, different kernels like BSD and what not. There are simply way too many moving parts and variables to be able to write software which works 100% correct all the time. Adrian -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 12/22/2017 04:00 AM, John Paul Adrian Glaubitz wrote:
On 12/22/2017 09:33 AM, Johannes Meixner wrote:
I think messages that are meant to be ignored by the user (i.e. messages that are meant only as optional information) should not be shown to the user by default.
Normally such messages may appear only in a log file or are shown to the user only in debug or verbose mode.
I don't know how excatly Ken Schneider ran 'zypper dup' i.e. with or without '--verbose' or '--details'.
Those messages usually show up as warning because it might not always be known from the programmer's point of view whether the reported issues might be a problem or not.
Warning basically says "I think this particular issue could cause an issue on your system but that's not 100% sure so I'll just inform you but let you continue anyway".
Cf. "WARNING is a waste of my time" at http://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html
Same here. Take gcc, for example. The compiler will warn you in many cases when you forget to initialize variables or perform a cast with incompatible data types which could result in your code crashing. However, since neither are a violation of the C language specification, the compiler will just let you continue. Sometimes the lack of initialization or the casts are intended by the programmer.
And in this case, Ken was also misreading the message. It clearly said there is no SysV script on the system, yet he assumed that was the case.
I made no such assumption. I wondered why 'any' SysV code would need to be run on a system without SysV installed. Plus, the
warning is actually a hint to the programmer that this particular piece of SysV code should be removed.
Better would be (if possible) for the packager to not have any SysV scripts be run at all if SysV is not installed. -- Ken Schneider -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Dec 22, 2017, at 5:44 PM, Ken Schneider - Factory <suse-list3@bout-tyme.net> wrote:
On 12/22/2017 04:00 AM, John Paul Adrian Glaubitz wrote:
On 12/22/2017 09:33 AM, Johannes Meixner wrote: I think messages that are meant to be ignored by the user (i.e. messages that are meant only as optional information) should not be shown to the user by default.
Normally such messages may appear only in a log file or are shown to the user only in debug or verbose mode.
I don't know how excatly Ken Schneider ran 'zypper dup' i.e. with or without '--verbose' or '--details'. Those messages usually show up as warning because it might not always be known from the programmer's point of view whether the reported issues might be a problem or not. Warning basically says "I think this particular issue could cause an issue on your system but that's not 100% sure so I'll just inform you but let you continue anyway". Cf. "WARNING is a waste of my time" at http://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html Same here. Take gcc, for example. The compiler will warn you in many cases when you forget to initialize variables or perform a cast with incompatible data types which could result in your code crashing. However, since neither are a violation of the C language specification, the compiler will just let you continue. Sometimes the lack of initialization or the casts are intended by the programmer. And in this case, Ken was also misreading the message. It clearly said there is no SysV script on the system, yet he assumed that was the case.
I made no such assumption. I wondered why 'any' SysV code would need to be run on a system without SysV installed.
I said „SysV script“, not „SysV“. And it didn’t run anything, it said „skipping“. You just misread the message. Adrian
Plus, the
warning is actually a hint to the programmer that this particular piece of SysV code should be removed.
Better would be (if possible) for the packager to not have any SysV scripts be run at all if SysV is not installed.
-- Ken Schneider -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Fri, Dec 22, Ken Schneider - Factory wrote:
I made no such assumption. I wondered why 'any' SysV code would need to be run on a system without SysV installed.
It does not try to run it, it looks during update, if there is a SysV init script from the old package, and if that configuration needs to be migrated to systemd. Make a fresh installation and you want see that message. Make a second update, and you should not see that message, too.
Better would be (if possible) for the packager to not have any SysV scripts be run at all if SysV is not installed.
Which is the case. Thorsten -- Thorsten Kukuk, Distinguished Engineer, Senior Architect SLES & CaaSP SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
23.12.2017 09:21, Thorsten Kukuk пишет:
On Fri, Dec 22, Ken Schneider - Factory wrote:
I made no such assumption. I wondered why 'any' SysV code would need to be run on a system without SysV installed.
It does not try to run it, it looks during update, if there is a SysV init script from the old package, and if that configuration needs to be migrated to systemd. Make a fresh installation and you want see that message.
Actually you should *not* see this message during fresh installation, it should simply skip this code. If not, something went wrong.
Make a second update, and you should not see that message, too.
Better would be (if possible) for the packager to not have any SysV scripts be run at all if SysV is not installed.
Which is the case.
Thorsten
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 22.12.2017 09:33, Johannes Meixner wrote:
Cf. "WARNING is a waste of my time" at http://blog.schlomo.schapiro.org/2015/04/warning-is-waste-of-my-time.html
But in this case, there was no "WARNING", it was just a message. It could still be just left off as it is useless IMHO anyway, but that's not my call ;) -- Stefan Seyfried "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Richard Feynman -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Fri, Dec 22, Stefan Seyfried wrote:
It could still be just left off as it is useless IMHO anyway, but that's not my call ;)
It's only useless for you as you don't debug migration issues. If you would do, you would not call it useless ;) -- Thorsten Kukuk, Distinguished Engineer, Senior Architect SLES & CaaSP SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 12/22/2017 03:33 AM, Johannes Meixner wrote:
Hello,
an addedum only FYI:
On Dec 21 19:24 John Paul Adrian Glaubitz wrote (excerpt):
On 12/21/2017 07:09 PM, Ken Schneider - openSUSE wrote:
Then what created this output:
SysV service man-db-create does not exist,
Looks like it clearly says something about a 'SysV service'
I think you're misinterpreting this message.
From your original post:
( 77/307) Installing: man-2.7.6-6.1.x86_64 ...........[done] Additional rpm output: SysV service man-db-create does not exist, skipping Updating /etc/sysconfig/cron ...
This means that the postinst script from the "man" package says that your system doesn't have the SysV service "man-db-create" which is why it's not trying to run the script in the first place.
This is just an information which you can ignore. The "man-db-create" SysV script doesn't exist on your machine because your system doesn't have SysVInit in the first place. It will just run the systemd equivalent command to create the database for manpages.
So, you can just ignore this message.
I think messages that are meant to be ignored by the user (i.e. messages that are meant only as optional information) should not be shown to the user by default.
Normally such messages may appear only in a log file or are shown to the user only in debug or verbose mode.
I don't know how excatly Ken Schneider ran 'zypper dup' i.e. with or without '--verbose' or '--details'.
I usually use 'zypper -vv dup'. Perhaps that is why I see the message. Thanks for the info, I will just ignore any messages about SysV in the future. -- Ken Schneider -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (8)
-
Andrei Borzenkov -
Cristian Rodríguez -
Johannes Meixner -
John Paul Adrian Glaubitz -
Ken Schneider - Factory -
Ken Schneider - openSUSE
-
Stefan Seyfried -
Thorsten Kukuk