Re: [CentOS] Performing post-installation setup tasks - for 75 minutes

2019-02-28 Thread Simon Matter via CentOS
> I'm doing a new install, and everything seems to have gone fine apart from
> the
> incredible amount of time it's taken.
>
> It does have 6 x  4TB drives in RAID6 and I am installing from a USB DVD
> drive
> as I had no SATA ports left, but this is getting rediculous.

Maybe you could try a network install and only use the USB DVD to boot. I
remember similar issues while installing servers using remote iso to
install via USB drive emulation with the servers management hardware.

Regards,
Simon

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


[CentOS-es] Falla de ssh

2019-02-28 Thread Roberto Bermúdez
Saludos amigos listeros, acudo a ustedes con un problema que es urgente
aunque paresca de risa, tengo un problema con un servidor el cual borré
todos los directoriso del directorio de root rm -vfr *, esto solo eliminó
los directorios no escondios, pero todos los que comienzan con punto "." no
fueron eliminados, el problema es que después de esta accion, ya no pude
ingresar a dicho servidor por medio de ssh (siempre uso la autenticación
por llave pública para no ingresar usaurio y contraseña), siempre desactivo
la atunticación por contraseña, pero con eso tuve que volver a activarlo
para no perder acceso a dicho servidor, por medio de este último métdo sí
puedo conectarme, en definitiva es que siento que el servidor ssh no lee el
archivo .ssh/authorized_keys, para verificar las llaves. Dentro del
/etc/ssh/sshd_config encuentro la línea #AuthorizedKeysFile
%h/.ssh/authorized_keys que siempre está cometnada, y aunque la descomente
el servidor no veirifica dicho archivo y por eso no puedo ingresar al
servidor

LO QUE YA HE HECHO
* regeneré la llave pública de ese servidor
* he descomentado la línea #AuthorizedKeysFile %h/.ssh/authorized_keys
en el archivo de configuración
* importé nuevamente la llave pública del equipo desde el cual quiero
conectarme
* eliminé el known_hosts del equipo cliente
* Ejecuté el comando yum --reinstall install  ssh-server
* Ejecuté el comando yum --reinstall install  ssh-import-id
Nada de esto me ha dado resultado, siempre me pide contraseña para ingresar

espero que alguien me pueda ayudar con este problema y agradezco de
antemano su gentil ayuda

Saludos fraternos
Roberto
___
CentOS-es mailing list
CentOS-es@centos.org
https://lists.centos.org/mailman/listinfo/centos-es


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread Jobst Schmalenbach
On Thu, Feb 28, 2019 at 05:19:49PM +0100, Nicolas Kovacs (i...@microlinux.fr) 
wrote:
> Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :
> > I want to lock in the SDA/SDB/SDC for my drives
> 
> In short : use UUIDs or labels instead of hardcoding /dev/sdX.

I **KNOW** how to use UUID's ... this is NOT the reason why I am doing this!


I *NEED* the order of the disks to be SDA(1st BIOS drive) SDB(2nd BIOS drive) 
SDC(3rd BIOS drive) and not SDA (1st BIOS drive) SDB(3rd BIOS drive) SDC (2nd 
BIOS drive).

Reason: it stuffs up the use of grub2* utilities leaving behind a bunch of 
error messages.
The SDA (1st BIOS drive) and SDB (2nd BIOS drive) are part of a MDADM raid(1) 
system.

As soon as I plug in the third drive, the OS (or systemD) decides to put it 
into the SDB spot - I do NOT want that.
When I the use any of the grub2 utils I end up with "missing drive" errors.

If I leave the drive out NO problem. I have managed twice to have the machine 
booting with the third drive as SDC, when that happens I I do not get any error 
messages.



Jobst


-- 
Why is the man who invests all your money called a broker?

  | |0| |   Jobst Schmalenbach, General Manager
  | | |0|   Barrett & Sales Essentials
  |0|0|0|   +61 3 9533 , POBox 277, Caulfield South, 3162, Australia
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread Jonathan Billings
On Feb 28, 2019, at 13:29, mark  wrote:
> No, I dislike UUIDs. I dislike, strongly, lots of extra typing that
> doesn't really get me anything. MAYBE, if you're in a Google or Amazon
> datacenter, with 500,000 physical servers (I phone interviewed with them
> 10 years ago)... but short of that? Nope.

I’ve never in my career ever had to type out a UUID. 

You don’t need to be that big to benefit from automation.  Even small shops 
would benefit from reproducible builds. Not every system needs to have 
loveingly crafted artisanal partition labels. 

All of this is moot, though, because I use lvm and so I just use 
/dev/volumegroup/logicalname, and that’s all assembled automatically in the 
kickstart. I only ever think about uuids when dealing with UEFI issues. 

—
Jonathan Billings 
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread mark
miguel medalha wrote:
>> No, I dislike UUIDs. I dislike, strongly, lots of extra typing that
>> doesn't really get me anything. MAYBE, if you're in a Google or Amazon
>> datacenter, with 500,000 physical servers (I phone interviewed with
>> them 10 years ago)... but short of that? Nope.
>>
> You can (perhaps should...) use the World Wide Name, which is a
> manufacturer ID unique to each disk. Contrary to the /sdX, it doesn't
> change with different configurations, OS or computer. An example of such
> an ID is the following:
>
> /dev/disk/by-id/wwn-0x50025ee3b4f5ca61
>
> Many modern disks have their WWN printed on their labels.
>
Why? And if I'm partitioning it, that won't work anyway. I partition, then
format with -l  and I don't *have* to change configuration, if
I'm say, replacing a failed disk. The labels I use *mean* something -
root, export, etc. Why would I want a meaningless id? That's like
companies who name everyone's computer some id, rather than, say,
mrothltp?

Hell, a few hours ago, a manager came to me to ask about network issues. I
thought I'd try to ping his system, and asked him the system name.  Of
*course* he couldn't remember it.

Self-documenting ia useful, if not carried overboard.

  mark


___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread miguel medalha

No, I dislike UUIDs. I dislike, strongly, lots of extra typing that
doesn't really get me anything. MAYBE, if you're in a Google or Amazon
datacenter, with 500,000 physical servers (I phone interviewed with them
10 years ago)... but short of that? Nope.


You can (perhaps should...) use the World Wide Name, which is a 
manufacturer ID unique to each disk. Contrary to the /sdX, it doesn't 
change with different configurations, OS or computer. An example of such 
an ID is the following:


/dev/disk/by-id/wwn-0x50025ee3b4f5ca61

Many modern disks have their WWN printed on their labels.

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread mark
Phelps, Matthew wrote:
> On Thu, Feb 28, 2019 at 11:52 AM mark  wrote:
>> Nicolas Kovacs wrote:
>>> Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :

 I want to lock in the SDA/SDB/SDC for my drives
>>>
>>> In short : use UUIDs or labels instead of hardcoding /dev/sdX.
>>>
>>> https://wiki.archlinux.org/index.php/persistent_block_device_naming
>>
>> Yeah - I strongly believe in labels, given the fact that *no* one can
>> remember a UUID
>
> ls -l /dev/disk/by-uuid
> (copy)
> (paste)

That is, of course, assuming you have a running system, that you haven't
replaced a hard drive, an MDraid's not giving you trouble, etc.

And oh, I put that in another system, and it's also got partition one
labled boot... so? I'm not trying to boot off of it, I'm going to mount it
on /mnt.

No, I dislike UUIDs. I dislike, strongly, lots of extra typing that
doesn't really get me anything. MAYBE, if you're in a Google or Amazon
datacenter, with 500,000 physical servers (I phone interviewed with them
10 years ago)... but short of that? Nope.

  mark

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Performing post-installation setup tasks - for 75 minutes

2019-02-28 Thread Gary Stainburn
On Thursday 28 February 2019 18:25:52 mark wrote:
> Gary Stainburn wrote:
> > I'm doing a new install, and everything seems to have gone fine apart
> > from the incredible amount of time it's taken.
> >
> > It does have 6 x  4TB drives in RAID6 and I am installing from a USB DVD
> > drive as I had no SATA ports left, but this is getting rediculous.
> >
> > I chose a simple install (the one above file and print server) and added
> > Postgresql but that was about it.
> >
> >
> > It took over 5 hours to do the install and at 16:50 it changed to
> >
> >
> > Performing post-installation setup tasks.  That was 85 minutes ago.  The
> > HDD
> > light keeps flashing, but other than that (and the circle swirling) there
> > doesn't seem to be any activity.
> >
> > Is this normal and can I check to see if it is actually doing something?
>
> That is seriously excessive, esp. since even if selinux is relabelling
> everything, it's a new system.
>
>  and see what journal or messages or dmesg have to say.
>
>   mark

dmesg has given lots of audit entries,

avc: denied {read} 

denied open and syslog reads etc too.  Bot if I repeatedly run dmesg I don't 
get new entries appearing.

/var/log/messages is empty

journalctl shows lots of 

systemd: 'cannot add depenency job for unit..' and 
kernel:  type=1400 audit avc: denied 

the denied entries are for getattr, execute, read, open, execute_no_trans, and 
setrlimit 
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Performing post-installation setup tasks - for 75 minutes

2019-02-28 Thread mark
Gary Stainburn wrote:
> I'm doing a new install, and everything seems to have gone fine apart
> from the incredible amount of time it's taken.
>
> It does have 6 x  4TB drives in RAID6 and I am installing from a USB DVD
> drive as I had no SATA ports left, but this is getting rediculous.
>
> I chose a simple install (the one above file and print server) and added
> Postgresql but that was about it.
>
>
> It took over 5 hours to do the install and at 16:50 it changed to
>
>
> Performing post-installation setup tasks.  That was 85 minutes ago.  The
> HDD
> light keeps flashing, but other than that (and the circle swirling) there
> doesn't seem to be any activity.
>
> Is this normal and can I check to see if it is actually doing something?

That is seriously excessive, esp. since even if selinux is relabelling
everything, it's a new system.

 and see what journal or messages or dmesg have to say.

  mark

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


[CentOS] Performing post-installation setup tasks - for 75 minutes

2019-02-28 Thread Gary Stainburn
I'm doing a new install, and everything seems to have gone fine apart from the 
incredible amount of time it's taken.

It does have 6 x  4TB drives in RAID6 and I am installing from a USB DVD drive 
as I had no SATA ports left, but this is getting rediculous.

I chose a simple install (the one above file and print server) and added 
Postgresql but that was about it.

It took over 5 hours to do the install and at 16:50 it changed to

Performing post-installation setup tasks.  That was 85 minutes ago.  The HDD 
light keeps flashing, but other than that (and the circle swirling) there 
doesn't seem to be any activity.

Is this normal and can I check to see if it is actually doing something?
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread Simon Matter via CentOS
> Nicolas Kovacs wrote:
>> Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :
>>
>>> I want to lock in the SDA/SDB/SDC for my drives
>>
>> In short : use UUIDs or labels instead of hardcoding /dev/sdX.
>>
>> https://wiki.archlinux.org/index.php/persistent_block_device_naming
>
> Yeah - I strongly believe in labels, given the fact that *no* one can
> remember a UUID

UUIDs can be helpful to automate things. On the other side they make
things more difficult than needed.

In my example I had two quite large servers with existing hardware. One of
the servers was already installed and in production. I had to clone the
hole system onto the second box but make sure that all UUIDs are unique in
the end.

Back in the old days that was an very easy task. But now, the hole system
consisted of 91 UUIDs for block devices/LVM/filesystems alone, with
additional UUIDs for other devices like ethernet interfaces.

I had a lot of fun!

Regards,
Simon

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread Valeri Galtsev



On 2/28/19 11:02 AM, Phelps, Matthew wrote:

On Thu, Feb 28, 2019 at 11:52 AM mark  wrote:


Nicolas Kovacs wrote:

Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :


I want to lock in the SDA/SDB/SDC for my drives


In short : use UUIDs or labels instead of hardcoding /dev/sdX.

https://wiki.archlinux.org/index.php/persistent_block_device_naming


Yeah - I strongly believe in labels, given the fact that *no* one can
remember a UUID

   mark



ls -l /dev/disk/by-uuid

(copy)

(paste)



I second Mark about filesystem labels. Labels you can read and type. 
UUIDs you can only copy and paste. Yes I did type them a few times, 
but... With upside there certainly comes downside of filesystem labels: 
if you are moving storage around you sometimes can hit the situation of 
having two identical labels. Which during last couple of decades I was 
able to get around.


Valeri






--

Valeri Galtsev
Sr System Administrator
Department of Astronomy and Astrophysics
Kavli Institute for Cosmological Physics
University of Chicago
Phone: 773-702-4247

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread Phelps, Matthew
On Thu, Feb 28, 2019 at 11:52 AM mark  wrote:

> Nicolas Kovacs wrote:
> > Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :
> >
> >> I want to lock in the SDA/SDB/SDC for my drives
> >
> > In short : use UUIDs or labels instead of hardcoding /dev/sdX.
> >
> > https://wiki.archlinux.org/index.php/persistent_block_device_naming
>
> Yeah - I strongly believe in labels, given the fact that *no* one can
> remember a UUID
>
>   mark
>

ls -l /dev/disk/by-uuid

(copy)

(paste)



-- 

*Matt Phelps*

*Information Technology Specialist, Systems Administrator*

(Computation Facility, Smithsonian Astrophysical Observatory)

Center for Astrophysics | Harvard & Smithsonian


60 Garden Street | MS 39 | Cambridge, MA 02138
email: mphe...@cfa.harvard.edu


cfa.harvard.edu | Facebook  | Twitter
 | YouTube 
| Newsletter 
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread mark
Nicolas Kovacs wrote:
> Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :
>
>> I want to lock in the SDA/SDB/SDC for my drives
>
> In short : use UUIDs or labels instead of hardcoding /dev/sdX.
>
> https://wiki.archlinux.org/index.php/persistent_block_device_naming

Yeah - I strongly believe in labels, given the fact that *no* one can
remember a UUID

  mark

___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] gpg?

2019-02-28 Thread Stephen John Smoogen
On Wed, 27 Feb 2019 at 19:39, mark  wrote:
>
> I just wanted to set up to send an encrypted message, so I went to generate a
> public/private key pair using gpg on C 6.
>
> Version 2.0.14, copyright 2009?
>
> Isn't there something newer than 10 years old?

Well CentOS-6 is nearing 8 years old.. and gpg is a core tool which
isn't going to get a lot of updates which might break RPM and other
tools which rely on the version that was done in 2009. So the versions
are going to be old. If you need some sort of newer feature you need
to compile a version for yourself in /usr/local or use a newer EL

>
> mark
> ___
> CentOS mailing list
> CentOS@centos.org
> https://lists.centos.org/mailman/listinfo/centos



-- 
Stephen J Smoogen.
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] What files to edit when changing the sdX of hard drives?

2019-02-28 Thread Nicolas Kovacs
Le 28/02/2019 à 04:12, Jobst Schmalenbach a écrit :
> I want to lock in the SDA/SDB/SDC for my drives

In short : use UUIDs or labels instead of hardcoding /dev/sdX.

https://wiki.archlinux.org/index.php/persistent_block_device_naming

Cheers,

Niki

-- 
Microlinux - Solutions informatiques durables
7, place de l'église - 30730 Montpezat
Site : https://www.microlinux.fr
Blog : https://blog.microlinux.fr
Mail : i...@microlinux.fr
Tél. : 04 66 63 10 32
___
CentOS mailing list
CentOS@centos.org
https://lists.centos.org/mailman/listinfo/centos


[CentOS-announce] CEBA-2019:C001 CentOS 7 shim BugFix Update

2019-02-28 Thread Johnny Hughes
CentOS Errata and Bugfix Advisory 2019:C001

This is a bug fix for for the following CentOS Bug:

https://bugs.centos.org/view.php?id=15522

The following updated files have been uploaded and are currently 
syncing to the mirrors: ( sha256sum Filename ) 


x86_64:
0ae55996308b507ba878377b35b71225c1fa28d4fb453c938a0d5dc9568501cc  
mokutil-15-2.el7.centos.x86_64.rpm
4e5cac6435cf60cd50610751e8dedc1df42d5b127a6aac033028d3dd049f01f6  
shim-ia32-15-2.el7.centos.x86_64.rpm
c8ba12bfa3c2cf5626e901e636536acba5c597a10d1081c0d5f780828c1a8a1d  
shim-unsigned-ia32-15-2.el7.centos.x86_64.rpm
affdaf712818dfdaa17ebbad94f08361bb318baf0db038c0131a10a4b2fa40c7  
shim-unsigned-x64-15-2.el7.centos.x86_64.rpm
8a8a958be8c80e53553199058f7612841e98f10bb1af4eb792f356353a6d985c  
shim-x64-15-2.el7.centos.x86_64.rpm

Source:
f951f93b68466e14772143f52f7903c9938b9914dc620cc96a45c8a1fb14ac16  
shim-15-2.el7.centos.src.rpm
67e0ba835f68457eae9b437199a13a1ad8c0b740e419824e5f2d813c025170fd  
shim-signed-15-2.el7.centos.src.rpm


-- Johnny Hughes CentOS Project { http://www.centos.org/ } irc:
hughesjr, #centos at irc.freenode.net
 Twitter:
@JohnnyCentOS




signature.asc
Description: OpenPGP digital signature
___
CentOS-announce mailing list
CentOS-announce@centos.org
https://lists.centos.org/mailman/listinfo/centos-announce