Discuss March 2015

discuss@lists.centos.org

172 participants
170 discussions

race conditions in dbus-1.2
by Danny Smit 31 Mar '15

31 Mar '15

Hi everyone, I'm having some issue with a Qt application using D-Bus (Qt API) in a threaded appllication on CentOS 6. Searching for the cause has led me to the following issues in the dbus library: https://bugs.freedesktop.org/show_bug.cgi?id=857 https://bugs.freedesktop.org/show_bug.cgi?id=17754 It appears that there are threading issues with the dbus library version < 1.6. However there doesn't seem to be other version than 1.2 available for CentOS 6. Can someone advise what approach can be used to either work around or solve the issue? For instance, is it even possible to upgrade to a newer dbus version without side effects? Kind regards, Danny Smit

1 0

updating tzdata on CentOS 7
by Dulmandakh 31 Mar '15

31 Mar '15

Hello, Government of Mongolia decided to adopt DST from this 28th March 2015. It’s changes are submitted and accepted by IANA. It seems that update to tzdata (2015b) issued for CentOS 6, but not for CentOS 7. Could you please tell when an update for CentOS 7 will be issued? Thank you.

3 3

Centos 6.4 on Xserve, IPMI error
by Arno B 31 Mar '15

31 Mar '15

Hello! I'm trying to install Centos 6.4 64 bit on this Xserve: http://www.everymac.com/systems/apple/xserve/specs/xserve-intel-xeon-2.8-ei… I got a bootloader from: http://blog.christophersmart.com/2009/07/23/linux-on-an-apple-xserve-efi-on… as none of the official Centos releases (full, minimal, network 64 bit) boot on this machine. They all crash when loading the bootx64.conf. I have built a USB stick with a working bootx64.efi, grub.cfg, together with an initrd and a vmlinuz from the isolinux folder. These seem to get me into booting a kernel until something called IPMI is starting to throw errors at me: Console: switching to colour frame buffer device 128x48 fb0: VESA VGA frame buffer device ipmi message handler version 39.2 IPMI System Interface driver. ipmi_si 0000:09:00.4: enable device (0002 -> 0003) ipmi_si 0000:09:00.4: PCI INT C -> GSI 16 (level, low) -> IRQ 16 ipmi_si: Adding PCI-specified kcs state machine ipmi_si 0000:09:00.7: enable device (0000 -> 0003) ipmi_si 0000:09:00.7: PCI INT D -> GSI 17 (level, low) -> IRQ 17 ipmi_si: Adding PCI-specified bt state machine ipmi_si: Trying PCI-specified kcs state machine at mem address 0x90841000, slave address 0x0 irq 16 IRQ 16/ipmi_si: IRQF_DISABLED is not guaranteed on shared IRQs Using irq 16 ipmi: Found new BMC (man_id: 0x00003f, prod_id: 0x0003, dev_id: 0x20) IPMI kcs interface initialized ipmi_si: Trying PCI-specified kcs state machine at mem address 0x90840000, slave address 0x0 irq 17 Refined TSC clocksource calibration: 2792.999 MHz. Switching to clocksource tsc IPMI BT: timeout in XACTION [ B_BUSY ] 1 retries left IPMI BT: timeout in XACTION [ B_BUSY ] failed 2 retries, sending error response IPMI: BT reset (takes 5 secs) IPMI BT: timeout in XACTION [ B_BUSY H2B ] failed 2 retries, sending error response IPMI: BT reset (takes 5 secs) IPMI BT: timeout in XACTION [ B_BUSY H2B ] failed 2 retries, sending error response IPMI: BT reset (takes 5 secs) IPMI BT: timeout in XACTION [ B_BUSY H2B ] failed 2 retries, sending error response IPMI: BT reset (takes 5 secs) IPMI BT: timeout in XACTION [ B_BUSY H2B ] failed 2 retries, sending error response IPMI: BT reset (takes 5 secs) then the screen goes to black and the machine doesn't respond to anything anymore... Can anyone please help me with this? Maybe just point me in a direction where to find a solution? I've been googling for a few days now without much results... thanks in advance Arno Beekman

2 1

viewvc for EL7?
by Les Mikesell 30 Mar '15

30 Mar '15

Is there a reliable repo with viewvc for Centos7? Or some reason it isn't in EPEL? -- Les Mikesell lesmikesell(a)gmail.com

1 0

db2 is running out of shared memory
by Harold Pritchett 30 Mar '15

30 Mar '15

Db2 is running our of shared memory. The environment: Virtual computers running under VMware ESXi hypervisor version 5.5. Two different virtual machines, each running CentOS 5, db2 express C, and Apache/PHP and vmware tools. The machines each have 8 GB of memory, 4 processors, and a couple of hundred GB of disk space available. Everything is running on the 64 bit versions. The actual hardware for VMware is a dual processor machine with two Xeon E5410 processors, each with four cpu cores and 32 GB of ram. The difference in the two virtual machines is that one is running db2 version 9.7.1 and the other is running db2 version 10.5.5. The error occurs while restoring a backup taken on another server to the test server in question. Worse than that, it occurs when restoring a second database after the first restore works OK. The error message we are getting is this: On the 9.7 system: RESTORE DATABASE XYZZY USER db2inst9 using FROM '/db2home/restore1' INTO XYZZZ REPLACE EXISTING WITHOUT ROLLING FORWARD WITHOUT PROMPTING SQL1084C Shared memory segments cannot be allocated. SQLSTATE=57019 On the 10.5 system: RESTORE DATABASE XYZZY USER db2inst9 using FROM '/db2home/restore1' INTO XYZZZ REPLACE EXISTING WITHOUT ROLLING FORWARD WITHOUT PROMPTING SQL1084C The database manager failed to allocate shared memory because an`operating system kernel memory limit has been reached. SQLSTATE=57019 Everything I have read online says to increas the value of Shared Memory Size in the /etc/sysctl.conf file. However, on these systems it is already set to a value eight times larger than the actual memory size. The command "ipcs -l" give: [root@cowtest etc]# ipcs -l ------ Shared Memory Limits -------- max number of segments = 4096 max seg size (kbytes) = 67108864 max total shared memory (kbytes) = 17179869184 min seg size (bytes) = 1 ------ Semaphore Limits -------- max number of arrays = 1024 max semaphores per array = 250 max semaphores system wide = 256000 max ops per semop call = 32 semaphore max value = 32767 ------ Messages: Limits -------- max queues system wide = 2048 max size of message (bytes) = 65536 default max size of queue (bytes) = 65536 [root@cowtest etc]# the file /etc/sysctl.conf has the following in it # Controls the maximum size of a message, in bytes kernel.msgmax = 65536 # Controls the maximum shared segment size, in bytes kernel.shmmax = 68719476736 # Controls the maximum number of shared memory segments, in pages kernel.shmall = 4294967296 These numbers just don't make sense. kernel.shmmax is set to 64 GB, 8 times the actual memory of the machine. the value of "Total Max Shared Memory" from the ipcs command is 16 TB! I have other machine running on real hardware with only 6 GB of memory on which this works fine. What is so strange about the virtual machines? I'm posting this to both the db2 and CentOS mailing lists. Sorry for the duplication if you are subscribed to both. Harold Attached below is the complete output of the backup.restore job which ran on the db2 10.5 machine. The backup was taken on a db2 9.7 machine, hence the message about database upgraded to current db2 version. backup.restore 1.0.0 - submitted Mon Mar 30 05:45:01 EDT 2015 on cowtest waiting... 1 05:45:01 waiting... 2 05:46:01 waiting... 3 05:47:01 waiting... 4 05:48:01 waiting... 5 05:49:03 backup.restore - started Mon Mar 30 05:50:03 EDT 2015 on cowtest File to be restored is: /db2home/www/daily/ANIMALS.0.db2inst9.NODE0000.CATN0000.20150330030114.001.bz2 Testing file to be restored for integrity at Mon Mar 30 05:50:03 EDT 2015 unzipping backup file ------------ /db2home/db2inst9/archivelogs/animals ------------ total 0 ------------ /db2home/db2inst9/archivelogs/animals ------------ SQL1064N DB2STOP processing was successful. Waiting for 1 minute... SQL1063N DB2START processing was successful. ATTACH TO db2inst9 Instance Attachment Information Instance server = DB2/LINUXX8664 10.5.5 Authorization ID = DB2INST9 Local instance alias = DB2INST9 DEACTIVATE DATABASE ANIMALS DB20000I The DEACTIVATE DATABASE command completed successfully. RESTORE DATABASE ANIMALS USER db2inst9 using FROM '/db2home/restore1' REPLACE EXISTING WITHOUT ROLLING FORWARD WITHOUT PROMPTING SQL2555I The database was restored and then successfully upgraded to the current DB2 release where you issued the RESTORE DATABASE command. CONNECT to animals user db2inst9 using Database Connection Information Database server = DB2/LINUXX8664 10.5.5 SQL authorization ID = DB2INST9 Local database alias = ANIMALS SELECT count(bhid) from is3.dna 1 ----------- 220 1 record(s) selected. DETACH DB20000I The DETACH command completed successfully. DISCONNECT CURRENT DB20000I The SQL DISCONNECT command completed successfully. ACTIVATE DATABASE ANIMALS DB20000I The ACTIVATE DATABASE command completed successfully. TERMINATE DB20000I The TERMINATE command completed successfully. ATTACH TO db2inst9 Instance Attachment Information Instance server = DB2/LINUXX8664 10.5.5 Authorization ID = DB2INST9 Local instance alias = DB2INST9 RESTORE DATABASE ANIMALS USER db2inst9 using FROM '/db2home/restore1' INTO VERIFY REPLACE EXISTING WITHOUT ROLLING FORWARD WITHOUT PROMPTING SQL1084C The database manager failed to allocate shared memory because an operating system kernel memory limit has been reached. SQLSTATE=57019 connect to verify user db2inst9 using SQL1084C The database manager failed to allocate shared memory because an operating system kernel memory limit has been reached. SQLSTATE=57019 select count(bhid) from is3.dna SQL1024N A database connection does not exist. SQLSTATE=08003 DETACH DB20000I The DETACH command completed successfully. TERMINATE DB20000I The TERMINATE command completed successfully. system is cowtest SQL1064N DB2STOP processing was successful. SQL1063N DB2START processing was successful. DB20000I The ACTIVATE DATABASE command completed successfully. ------------ /db2home/db2inst9/archivelogs/20150330 ------------ total 0 ------------ /db2home/db2inst9/archivelogs/20150330 ------------ backup.complete file has already been removed... backup.restore - ended Mon Mar 30 07:44:52 EDT 2015 on cowtest

3 3

mysql can't connect from localhost -strange behavior
by Tim Dunphy 30 Mar '15

30 Mar '15

Hey all, I've been having some trouble creating a mysql user that can connect to the database from localhost. It's always been a straight forward thing to do in the past, so its time for a sanity check, if you guys don't mind. Ok, so here's the actual command with actual simplified password that I'm using. It's on localhost so I don't think it's a security threat unless someone gets access to the box itself. mysql> grant all privileges on ftp.* to 'proftpd'@'localhost' identified by 'testpattern'; Query OK, 0 rows affected (0.35 sec) Here's what it looks like when you select it from the mysql database: mysql> select User,Host,Password from user where User like 'proftpd'; +---------+-----------+-------------------------------------------+ | User | Host | Password | +---------+-----------+-------------------------------------------+ | proftpd | localhost | *2EE931CA39652F1ED359A3A36961511B387E74A9 | +---------+-----------+-------------------------------------------+ 1 row in set (0.00 sec) And here's my attempt to connect with the password shown. Which is something I don't usually do, but am doing now to demonstrate what's going on: [root@ops:~] #mysql -uproftpd -ptestpattern -h localhost ERROR 1045 (28000): Access denied for user 'proftpd'@'localhost' (using password: YES) OK, so as I've said this should work! The database I'm trying to give the user access to does also exist: mysql> show databases like 'ftp'; +----------------+ | Database (ftp) | +----------------+ | ftp | +----------------+ 1 row in set (0.34 sec) I checked the error log for mysql and didn't find any clues there: [root@ops:~] #grep log /etc/my.cnf log-error=/var/log/mysqld.log [root@ops:~] #tail /var/log/mysqld.log InnoDB: Restoring possible half-written data pages from the doublewrite InnoDB: buffer... 150329 13:30:34 InnoDB: Waiting for the background threads to start 150329 13:30:35 InnoDB: 5.5.42 started; log sequence number 6071094973 150329 13:30:35 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306 150329 13:30:35 [Note] - '0.0.0.0' resolves to '0.0.0.0'; 150329 13:30:35 [Note] Server socket created on IP: '0.0.0.0'. 150329 13:30:35 [Note] Event Scheduler: Loaded 0 events 150329 13:30:35 [Note] /usr/libexec/mysqld: ready for connections. Version: '5.5.42' socket: '/var/lib/mysql/mysql.sock' port: 3306 MySQL Community Server (GPL) by Remi Does anybody out there have any idea why this isn't working? Thanks Tim -- GPG me!! gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B

5 6

headless VirtualBox on Centos
by Boris Epstein 30 Mar '15

30 Mar '15

Hello listmates, I am wondering if there is a set of scripts/utilities for automatically starting and running headless (no X11) VM's using VirtualBox omn a CentOS 6 server. VNC/RDP access to the VM's would be fine. Any help much appreciated. Cheers, Boris.

7 10

RAID1 bootloader configuration on CentOS 6.x and 7
by Niki Kovacs 29 Mar '15

29 Mar '15

Hi, The CentOS wiki sports a page about setting up software RAID1 on CentOS 5.x. There's a section about making both members of the RAID1 bootable by setting up GRUB on both disks. Now I wonder how this should be done on CentOS 6.x and 7. I have two sandbox machines in my office, one running a minimal CentOS 6.6, the other one with a CentOS 7 installation. Correct me if I'm wrong, but I guess the setup on CentOS 6.6 is similar to the one described on the wiki page for CentOS 5.x, something like: grub> device (hd0) /dev/sda grub> device (hd1) /dev/sdb grub> root (hd0,0) grub> setup (hd0) grub> root (hd1,0) grub> setup (hd1) grub> quit Now how would that work with the new GRUB2 under CentOS 7? Or maybe it's already installed on both disks, but how would I know that? Cheers, Niki -- Microlinux - Solutions informatiques 100% Linux et logiciels libres 7, place de l'église - 30730 Montpezat Web : http://www.microlinux.fr Mail : info(a)microlinux.fr Tél. : 04 66 63 10 32

2 1

Why is irqbalance not balancing?
by Steve Snyder 29 Mar '15

29 Mar '15

I am running irqbalance with default configuration on an Atom 330 machine. This CPU has 2 physical cores + 2 SMT (aka Hyperthreading) cores. As shown below the interrupt for the eth0 device is always on CPUs 0 and 1, with CPUs 2 and 3 left idle. But why? Maybe irqbalance prefers physical cores? My understanding, though, is that the even-numbered CPUs are the physical cores, with the odd-numbered one being the SMT cores. If this understanding is correct, it means that irqbalance is toggling between a single physical core and its SMT sibling. Any thoughts on why irqbalance is not using all 4 CPUs to distribute the eth0 interrupts? Thanks. --------------------- # cat /proc/interrupts CPU0 CPU1 CPU2 CPU3 0: 717 568 0 0 IO-APIC-edge timer 1: 0 2 0 0 IO-APIC-edge i8042 8: 26 33 0 0 IO-APIC-edge rtc0 9: 0 0 0 0 IO-APIC-fasteoi acpi 12: 1 3 0 0 IO-APIC-edge i8042 17: 11409 11661 0 0 IO-APIC-fasteoi ahci 18: 264 424 0 0 IO-APIC-fasteoi snd_hda_intel 20: 0 0 0 0 IO-APIC-fasteoi ohci_hcd:usb2 21: 0 0 0 0 IO-APIC-fasteoi ohci_hcd:usb3 22: 0 0 0 0 IO-APIC-fasteoi ehci_hcd:usb1 24: 686292451 685980205 0 0 PCI-MSI-edge eth0

3 2

reboot causes mapping failures after update to Centos 7
by Stephen 29 Mar '15

29 Mar '15

Hi, I perform the initial update of about 200 packages then when I reboot I get three options instead of two. One is the Core 7 and the other is x86 what do those mean? Also after this the journal returns mapping errors and ext4 errors Best, Steve

1 0

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss March 2015