similar to: Strange Performance Issue / concurrent clients -> very very slow

Hi, I'm looking for some recommendations for a "fast" fileserver regarding the hardware you use. We have different fileservers as our requirements changed over time. The "main" problem we are faced with is, that with smb (windows 7 and OS X) clients we never get really close to GBit speed on reads or writes. Using the same servers/storages with ftp, ssh, rsync, nfs we

Hi, I do have one iscsi storage with 4 GBit nics of which currently only one is configured with an ip and which is in productive use by one cent os 6.3 server. Doing some research brought me to the idea to use some nic bonding or multipathing for that storage, but multipathing I did about four or five years ago only once :) Furthormore I can't find the ultimate answer (may be there is not

Hi, I install a new 4 Port GBit Server with centos 6.3 and I'd like to use nic bonding. Searching for a good how to pinted to this page: http://www.how2centos.com/centos-6-channel-bonding/ My question: Is this the current, good way to bond the 4 ports? The other question: What should I set on the cisco switch for the Port-channel options? As far as I understand for performance and

Hi, I do have one fileserver, which is compared to other systems slow when transferring file through the lan. The test I did get a transfer speed of about 70 - 80 MB/s. I copy files by rsync or scp from a SSD disk to the servers ISCSI Storage and I tested to copy to a serverside RAM disk. I copy one big file of about 2/10 GB; 2 is to the RAM disk. The point is, the mentioned one server just

Hi, thanks for all the feedback and topics. May be I can/should draw a better picture of my settings and question. We "still" use samba 3.6.9-167 from Red Hat with an openldap backend. I tried from an older posting somewhere: socket options = TCP_NODELAY IPTOS_LOWDELAY SO_RCVBUF=65536 SO_SNDBUF=65536 But there was no noticeable impact. The log level = 2. Thats all I changed

Hi, recently I got my samba/ldap/pdc setup up and running with xp clients. Server: centos 5.5, openldap-2.3.43-12 and samba3x-3.3.8. The new challenge: adding Windows 7 Clients and upgrading to samba 3.3.x I followed a posting from the centos forum (1) which worked so far for the samba update breaking nothing for xp. Than I installed a windows 7 client with the latest patches, changed the

Hi, currently we do have one samba3x-3.5.10-0.109.el5_8 RH EL 5.8 PDC authenticating by our central LDAP server. This PDS also hosts the central fileserver storage for all our +- 600 users, some group shares and roaming profiles. The clients are OS X, Win XP and Win 7. We hope to have all XP 'killed' by end of the year. Furthermore we do have a second stand alone samba server for some

Hi, did anybody did an update from 1.x -> 2.x yet? I'D like to use the rpm from http://packages.atrpms.net/dist/el5/dovecot/ May be someone has some suggestions? Regards . G?tz -- G?tz Reinicke IT-Koordinator Tel. +49 7141 969 420 Fax +49 7141 969 55 420 E-Mail goetz.reinicke at filmakademie.de Filmakademie Baden-W?rttemberg GmbH Akademiehof 10 71638 Ludwigsburg www.filmakademie.de

Hi, one more LVM thing: on one server some LVM stores are physically removed but still there are some informatione somwher spread adcross teh server. e.g. if I do a vgdisplay or pvcreat I get an error regarding that old storage: /dev/raid_10/lvol0: read failed after 0 of 4096 at 14000515383296: Eingabe-/Ausgabefehler How can I remove all information? Thanks for any hint. Regards . G?tz

Hi, I do have an iscsi storage with two raidsets. I'm logged in to the target and get two devices: /dev/sdb and /dev/sdc. After formatting the devices with ext4 I realised, that I had used a wrong label, so I tried to relabel the devices. But now I'm confused: doing a e2label /dev/sdb or /dev/sdc returns always the same label name for both devices. so I did "e2label /dev/sdb

Hi, we still run dovecot 1.2.17 and upgrading is not planed for some time. But I'm asked to install a sieve system to be used with SOGo soon. What do I have to install / configure? Is installing/configuring the dovecot-sieve-0.1.19 and dovecot-managesieve-0.11.13 enough? We still use mostly the default settings for imaps/pop3s. And of course configure SOGo. Thanks for any feedback.

Hi, I do have two DNS servers running bind 9.7.3. One master, one slave. If I add/update a record on the master, it takes up to more or less 20 hours until that change is transferred to the slave. That is a long time for me :-) May be I got some settings wrong which would sync much faster ... ? What could be wrong? What might I check? Thanks for any suggestion and hint! Regards . G?tz --

Hi, some time ago I removed some physical disks from a server and now I'm still getting dmesg messages like: sd 0:2:2:0: SCSI error: return code = 0x00040001 end_request: I/O error, dev sdc, sector 0 And all lvm tools still grumbel about that disks too: /dev/sdb: read failed after 0 of 4096 at 0: Eingabe-/Ausgabefehler /dev/sdb: read failed after 0 of 4096 at 1746969493504:

Hi, it's time to start our redundant dhcp server setup and I found some suggestions and tutorials. But all setups I found use either static or nonstatic ip adress management. For the static setup I'd setup two servers and copy the configfile from the 'master' to the secondary server and do a dhcpd restart. For the nonstatic I have to setup pool and failover peer configurations

Thank you for clarifying this, Johan. Very much appreciated! On Wed, 29 Apr 2015 22:28:00 +0200 Johan Kooijman wrote: > Carl, > > By default my.cnf has to obey the OS limits, so in this case the > order is: systemd > /etc/security/limits* > /etc/my*. > > On Wed, Apr 29, 2015 at 3:22 PM, Carl E. Hartung > <carlh04426 at gmail.com> wrote: > > > Hi

Hi, in one server I do have a SSD raid 1 size 219GB. df shows 9.4 GB free, 198GB used. If I do "du -sch * | sort -h -r" on /, I just have close to 3.5GB used .... Any hints, what's eating up the space? Centos 6.6, fs = ext4. regards . G?tz -- G?tz Reinicke IT-Koordinator Tel. +49 7141 969 82 420 E-Mail goetz.reinicke at filmakademie.de Filmakademie Baden-W?rttemberg GmbH

Gotz, This is due to systemd, it overrules your settings. Add a file to systemd config fixes it: [root at mysql2 ~]# cat /etc/systemd/system/mariadb.service.d/limits.conf [Service] LimitNOFILE=10000 LimitMEMLOCK=10000 On Wed, Apr 29, 2015 at 8:31 AM, G?tz Reinicke - IT Koordinator < goetz.reinicke at filmakademie.de> wrote: > Hi, > > may be somewon has a working solution and

Carl, By default my.cnf has to obey the OS limits, so in this case the order is: systemd > /etc/security/limits* > /etc/my*. On Wed, Apr 29, 2015 at 3:22 PM, Carl E. Hartung <carlh04426 at gmail.com> wrote: > Hi Johan, > > Does systemd also overrule /etc/my.conf? > > Thx! > > Carl > > On Wed, 29 Apr 2015 14:58:52 +0200 > Johan Kooijman wrote: > >

Usually problem in encryption. try cipher arcfour or apply hpn patches to ssh. ( http://www.psc.edu/index.php/hpn-ssh) -- Eero 2015-08-11 12:37 GMT+03:00 G?tz Reinicke - IT Koordinator < goetz.reinicke at filmakademie.de>: > Hi, > > i have two servers, connected to to the lan by 10Gb with 10Gb and DAS > hardware raid. > > Each system con read and write locally or to the

Hi, I do a tgz-backup some maildir-folders with n*1000 off files and a lot of GB in storage. The backuped maildirs are removed after the tar. My assumption was, that the free diskspace should be bigger after that, but from what I get with df, it looks like I'm loosing space. Currently the tgz is saved on the same disk/mountpoint. Any hint, why removing the maildirs dont free diskspace as