Postings by lejeczek

shellinabox via proxy(apache)

hi guys,

cannot get it to work - shellinabox - not being programmer nor
selinux sorcerer.

shellinabox via apache, when I ausearch it all I get is:

#============= unconfined_service_t ==============

#!!!! The file '/usr/bin/bash' is mislabeled on your system.

rpmbuild --define | some rpm sorcerer around?


how do you pass vars to rpmbuild for definition? eg

rpmbuild --define \'"${_definition2}"\'

I've been fiddling with ways to escape, but none is fricking working..
I mean, rpmbuild rushes to work(no errors nor failure) so if you try
just the command line do not believe it, because later as it executes
%if you will see process does not see these definitions.

many thanks, L

can you rpm rebuild R-core-3.5.0-1.el7.x86_64 from epel-testing..


many thanks, L.

FreeIPA - client/replica errors

hi, free IPA everyone?

I wanted to ask if you maybe seen below errors.

systemd - service restart another service... (how?)

... but only upon its own start/restart

hi everyone

would you know a trick?
PartOf= does the whole lot, stops when stopped, so no.
Here I need only re/start, everything else is ignored.

many thanks, L.

AMD epyc/naples

do we know if 7.4 is good for new AMD cpus?
More specifically for virtualization. Historically there
were some issues and things like AMD's IOMMU & similar were
not exactly perfectly supported by HW vendors, UEFI/bios <=>

I'm thinking of getting one and I wonder if it might be too
soon. I wonder if above things are at least in an acceptable
shape of older Opteron platform, and not worse. Any experience?

b.w. L.

lvconvert(split) - raid10 => raid0

hi guys, gals

do you know if conversion from lvm's raid10 to raid0 is
I'm fiddling with --splitmirrors but it gets me nowhere.
On "takeover" subject man pages says: "..between
striped/raid0 and raid10."" but no details, nowhere I could
find documentation, nor a howto.

many thanks, L.

Seagate's firmware flash until in Centos 7.x - does it work?

hi fellas

I wonder if any of you use(d) dl_sea_fw to flash SAS drive?
I try dl_sea_fw-0.2.3_64 to flash ST32000444SS but it fails:

$ ./linux\ cli\ tools/dl_sea_fw-0.2.3_64 -m ST32000444SS -f
firmware/MU-SAS-0008.LOD -d /dev/sg59
 Seagate Firmware Download Utility v0.2.3 Build Date: Jan 
9 2013
 Copyright (c) 2012 Seagate Technology LLC, All Rights Reserved
 Thu Oct 12 11:33:05 2017
Downloading file firmware/MU-SAS-0008.LOD to /

owncloud - external storage - smb

hi fellas

does anybody use by a chance external smb storage in current
owncloud version?
If yes does it work for you?

I have, otherwise fully OK samba, but my owncload cannot get
to it.
On my setup(pretty default) users log into owncloud with an
ldap backend, the very same samba uses, yet in samba logs I see:

2017/10/05 15:41:00.479585,  5]
  make_server_info_sam: made server info for user
cloudadmin -> cloudadmin
[2017/10/05 15:41:00.479709,  3]

fetch to and forward on one server - how?

hi fellas

Before I start tampering with stuff I thought - better ask

Having one postfix box which for local users forward to a
remote/external box(probably postfix too) - can I have that
same my postfix box fetch from that external server?
Would I have to use fetchmail(or similar) or postfix could
do it itself?

What I'm hoping is that some expert could say if there is a
potential to cause a disaster like some loop where fetchmail
or postfix will be retrieving from remote and also, by
misconfiguration, by misuse will be forwarding that
retrieved emails again to external serv

exclude a host(s) and allow it without authentication

hi everyone

I'm fiddling with settings but thought, someone already must
know - how to achieve above, if possible at all?

many thanks,

have udev to ignore/hide a scsi disk - but how?

hi everyone

I've been playing and trying to tell udev to ignore and not
create symlinks in /dev but I fail.
Hoping this is the one: ENV{ID_SERIAL}== I've tried:

but udev tests would still show: ACTION=add and symlinks

Would share an advice?
many thanks,

that ever puzzling special chars escaping + rdiff-backup

hi everyone

a good basher around here?

I try in a script:

_rdiffBack="rdiff-backup -v5 --tempdir /tmp/ --no-eas
--exclude-other-filesystems --exclude-symbolic-links"
_rdffiExclude="--exclude '**/~*' --exclude '**.tmp'
--exclude-regexp '(.glusterfs|.trashcan|temp)'"

__backMeUp() {
for _sourceDir in ${_sourceDirs[@]}; do
echo \$ ${_execCom} ${_rdffiExclude} ${_sourceDir}/
${_backupTo} \> ${_logFile} | ${_copyToLog}
${_execCom} ${_rdffiExclude} ${_sourceDir}/
${_backupTo} > ${_logFile} 2>&1

case ${1} in

raid 10 not in consistent state?

hi everyone
I've just configured a simple raid10 on a Dell system, but
one thing is puzzling to me.
I'm seeing this below and I wonder why?

weird (non-functional) downgrade

dear devel

is this true that every rpm downgrade goes kind of a ...

samba rpm deps - with yum downgrade = kind of a mayhem

hi guys, gals

do you see this:

~]$ yum downgrade samba
Resolving Dependencies
--> Running transaction check
---> Package samba.x86_64 0:4.4.4-9.el7 will be a downgrade
--> Processing Dependency: samba-libs = 4.4.4-9.el7 for
package: samba-4.4.4-9.el7.x86_64
--> Processing Dependency: samba-common-tools = 4.4.4-9.el7
for package: samba-4.4.4-9.el7.x86_64
--> Processing Dependency: samba-common-libs = 4.4.4-9.el7
for package: samba-4.4.4-9.el7.x86_64
--> Processing Dependency: samba-common = 4.4.4-9.el7 for
package: samba-4.4.4-9.el7.x86_64
--> Processing Dependency: samba-client-libs = 4.

does your kdump work?

hi all

mine does not, and I've tampered with it in many ways,
cannot get it to work.


devtoolset-4 ageing?

hi everyone

just a quickie to devel maybe. I'm looking at some bits:

devtoolset-4-elfutils-libelf-0.163-2.el7.x86_64 VS

isn't devtool ageing? Could this be the case with more packages?

ipa rpm conflicts

does anybody see this:

unning transaction
Warning: RPMDB altered outside of yum.
** Found 8 pre-existing rpmdb problem(s), 'yum check' output
ipa-admintools-4.4.0-14.el7.centos.noarch has installed
conflicts freeipa-admintools:
ipa-client-4.4.0-14.el7.centos.x86_64 has installed
conflicts freeipa-client: ipa-client-4.4.0-14.el7.centos.x86_64
ipa-client-common-4.4.0-14.el7.centos.noarch has installed
conflicts freeipa-client-common:
ipa-common-4.4.0-14.el7.centos.noarch has installed

something going on with CR ?

I get:

Failed to download prestodelta for repository cr: [Errno
256] No more mirrors to try.
java-1.8.0-openjdk- FAILED
<a href="" title=""></a>
[Errno 14] HTTP Error 404 - Not Found
Trying other mirror.
java-1.8.0-openjdk-devel-1.8.0 FAILED
<a href="" title=""></a>
[Errno 14] HTTP Error 404 - Not Found
Trying other mirror.

glusterfs bits..

hi everyone

before I subscribe to yet another @list I was hoping a
centosian here has some experience and know the answer -
I've started googling but have to failed - glusterfs does
not translate local FS (xfs) quotas and presents them to its
clients, does it?
It is only glusterfs own quota functions that we have to
manage quoatas?

many thanks.

samba CTDB inability to create smb.conf cache

hi everyone

any experience with ctdb?
I'm setting up a simple cluster, only ldap backend instead
of tdb2.
One (of the two) server fails this way:

50.samba: ERROR: smb.conf cache create failed

$ ctdb status
Number of nodes:2
pnn:0 OK
hash:0 lmaster:0
hash:1 lmaster:1
Recovery mode:NORMAL (0)
Recovery master:0

If I copy the file manually while ctdb is running I see:

50.samba: WARNING: smb.conf cache update failed - using old
cache file
Node became HEALTHY.

dnf and failing epel

hi everyone,

I use dnf but one epel fails in a peculiar way:

$ dnf repolist
Failed to open:

that file is there, so I tampered with perms & fcontext but
to no avail. dnf clean does remove it but on the first
contact to epel repo it fails again.
Do you see it with your centoses?


firewalld and just a routing between local lans

how much different iptables are in 7 from 6 I'm just
learning, I expected that since IFs belong to the same zone
traffic between them would pass, be forwarded without any
extra settings/configuration but, I'm wrong it turns out.
Am I missing something and there actually is a simple way,
designed to make that routing between lans/interfaces just
work - a one liner, one option/setting in that
comprehensive(?) firewalld configuration?
How does firewall make your centos a router of directly
connected LANs?

many thanks

local routing puzzle

hi everyone

I'm having a bit of a puzzle on one centos system, I have
four interfaces:

I do:
ping -I em1
PING ( from em1: 56(84)
bytes of data.
64 bytes from icmp_seq=1 ttl=64 time=0.367 ms

the same for -I em2, for -I em1 and naturally for -I
nm-team1 - all good.
$ ping -I p3p3
PING ( from p3p3: 56(84)
bytes of data.

and nothing, ping waits and no reply, Ctrl+C

with such a simple setup rules based routing should

find any row with NULL

hi there,

I've been searching the vastness of the net but cannot find
- how - to get all the rows with a NULL.

And like any novice I wonder if this can be done without
reiterating all the columns names(not manually at least)?

Some expert would say it is easy, how easy is it? with an

many thanks


I wonder if I get this email...

hi everybody.

I'm a subscriber to a few mailing lists, for a good number
of years now, but recently something very annoying happened
and I'm hoping postfix list here there is someone could
suggest some troubleshooting that may resolve this problem.

I use Yahoo free of charge mail service, about two weeks ago
I stopped getting my own messages I send to any list.

I see they go through for people reply and I get the replies.

I checked all the filters, both server and local side,
nothing, but I did not change anything there around when it
happened anyway.

To eliminate third parties I tr

virtually identical support but..

hi everybody

there is a but :) I fail to understand,
I have two virtually identical setups, both relay to one
specific domain, yet! - one gets: 550-Verification failed
550-Invalid domain part in email address.

gee... what am I missing?
Both servers are on the same net, and the rest of relevant
(that I could think of) parts should be the same.

for ideas & suggestions,
many thanks,

failed to decrypt log block

hi everybody..

would you know how to recover from such a problem:

InnoDB: ok header, but checksum field contains 792537472,
should be 1776874443
2016-04-04 12:41:15 140333716928640 [ERROR] InnoDB: Redo log
crypto: failed to decrypt log block.