It seems not be a problem with your configuration if you strictly follow the Installation Manual. It looks like ceph was successfully downloaded the last time you tried. I guess it might be some issue with your virtual machine's configuration. Could you try to deploy a cluster without virtual machine? Or in another network environment? What is your VM's network configuration?
Zhe From: Yuriy Grabar [mailto:[email protected]] Sent: Tuesday, June 17, 2014 2:46 AM To: Zhe Zhang; [email protected] Cc: Vladyslav Dehtiar -X (vdehtiar - SOFTSERVE INC at Cisco) ([email protected]); [email protected] Subject: RE: [ceph-users] Assistance in deploying Ceph cluster Hi Zhe, Please find attached output. We use https proxy server in our Lab. It seems that 'deploy script' cannot get access to https://ceph.com/git/?p=ceph.git;a=blob_plain;f=keys/release.asc I set the next string (export https_proxy=http://proxy-wsa.esl.cisco.com:80) to ~/.bash_profile and it didn't help Regards Yuriy From: Zhe Zhang [mailto:[email protected]] Sent: Tuesday, June 17, 2014 2:16 AM To: Yuriy Grabar; [email protected]<mailto:[email protected]> Cc: Vladyslav Dehtiar -X (vdehtiar - SOFTSERVE INC at Cisco) ([email protected]<mailto:[email protected]>); [email protected]<mailto:[email protected]> Subject: RE: [ceph-users] Assistance in deploying Ceph cluster What is your "hostname -s" ? In ~/.ssh/config file, you have to configure the hostname the same as what shows with "hostname -s" Zhe From: ceph-users [mailto:[email protected]] On Behalf Of Yuriy Grabar Sent: Monday, June 16, 2014 2:41 AM To: [email protected]<mailto:[email protected]> Cc: Vladyslav Dehtiar -X (vdehtiar - SOFTSERVE INC at Cisco) ([email protected]<mailto:[email protected]>); [email protected]<mailto:[email protected]> Subject: [ceph-users] Assistance in deploying Ceph cluster Hello everyone, We are looking for support in installing Ceph cluster (1 admin node + 3 node, based on Centos 6.5 Vsphere Vm's) We observe some errors in executing ceph-deploy commands. (all keys and permissions are configured correctly) There is attached log of execution for "ceph-deploy install node1" command. Seems that curl script tries to connect using ipv6 address. Node1 has only one network interface. Note: We had tried to disable ipv6 protocol on node1 side, anyway we faced error like "host not found" According attached log, We understand that problem is in importing rpm key. If I run 'rmp import' manually, it executes on Node1 successfully. Also we know that instead of using 'ceph-deploy install' command we can manually install Ceph service and related tools on Node1 side, but would we like to follow guidance on official site (http://ceph.com/docs/master/start/quick-ceph-deploy/) as it was developed for that. Please suggest me what could be wrong in configuration from my side. Thanks and Regards, Yuriy Hrabar Quality Control engineer [cid:[email protected]] 52 V.Velykoho, Lviv 79053, Ukraine Tel: +380-32-240-9090 x 3290 Cell: +380-50-430-3066 Fax: +380-32-240-9080
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
