Mailing-List: contact users-help@cloudstack.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@cloudstack.apache.org
Received-SPF: pass (athena.apache.org: domain of iyoung@ratespecial.com
 designates 209.85.223.182 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAO0CpcTD+-OLRkHjhw_ZOsV6wSn3pzzPLe=wsoL8bmr1HLzHjQ@mail.gmail.com>
References: 
 <CAO0CpcR=s8rHmmOq0jjaA7647ZKO4awvfqRSFZ82NjHzBUpmOw@mail.gmail.com>
	<CAO0CpcTj_2BNarmXk_24q5qkCFWPv-24apuTQkj5s221dKAWKw@mail.gmail.com>
	<CAO0CpcQtVreWPnAv=hBdfLbeCC3Z1irUt+aBZxuG0sEPNaiRdA@mail.gmail.com>
	<CAO0CpcTD+-OLRkHjhw_ZOsV6wSn3pzzPLe=wsoL8bmr1HLzHjQ@mail.gmail.com>
Date: Mon, 12 May 2014 15:40:19 -0700
Message-ID: 
 <CAO0CpcTakrveYsZcmX4nkRU5Tq3=xt3h7skY_M=z9JQcfeB_gg@mail.gmail.com>
Subject: Re: new installation--ssvm won't start
From: Ian Young <iyoung@ratespecial.com>
To: users@cloudstack.apache.org
Content-Type: multipart/alternative; boundary=089e0115ff70c7609e04f93ba1ce

--089e0115ff70c7609e04f93ba1ce
Content-Type: text/plain; charset=UTF-8

I was able to complete the installation on Friday.  Two things I did
differently that were not mentioned in the quick start guide were to
disable requiretty in /etc/sudoers and to set up NFSv4 correctly (i.e. set
up a global root directory with fsid=0).  I'm not sure how much impact the
sudoers configuration had on my problem but I'm pretty sure the NFS setup
was the main issue.


On Thu, May 8, 2014 at 5:33 PM, Ian Young <iyoung@ratespecial.com> wrote:

> I know this has something to do with idmapd and NFS.  This error keeps
> appearing in /var/log/messages:
> May  8 10:29:54 virthost1 rpc.idmapd[11044]: nss_getpwnam: name '0' does
> not map into domain 'redacted.com'
>
>
> On Thu, May 8, 2014 at 5:20 PM, Ian Young <iyoung@ratespecial.com> wrote:
>
>> I wiped the server clean and started over again today.  In the process, I
>> realized that, the previous time, I forgot to uncomment the Domain line in
>> /etc/idmapd.conf.  However, even though I included the step this time, the
>> GUI installer still seems to hang on the final "Creating system VMs" step.
>>  I see two VMs running when I run "virsh list" (the secondary storage VM
>> keeps getting regenerated).  In the primary storage, it looks like there is
>> one complete 693 MB image but the other two are only 11 and 12 MB, although
>> they are gradually growing.  What's happening here?
>>
>> [root@virthost1 ~]# ls -hl /var/primary/
>> total 715M
>> -rwxr--r--. 1 nobody nobody  11M May  8 09:55
>> 54de167f-ad9c-453b-91c7-fdd644922932
>> -rwxr--r--. 1 nobody nobody  12M May  8 09:55
>> 91069b66-b1b3-41aa-8995-874fd4353473
>> -rwxr--r--. 1 nobody nobody 693M May  8 09:16
>> c2e6efba-d6c7-11e3-9e76-002590c96d30
>>
>> The management server log keeps reporting that "There is no secondary
>> storage VM for secondary storage host nfs://192.168.100.6/var/secondary."
>>  Here is a larger section of logs:
>> http://pastebin.com/NFf5cBx3
>>
>>
>> On Wed, May 7, 2014 at 10:49 AM, Ian Young <iyoung@ratespecial.com>wrote:
>>
>>> I noticed that in Home > Infrastructure > Zones > Zone1, Resources tab,
>>> the Secondary Storage says "Allocated 0.00 KB / 0.00 KB".  However, the
>>> secondary storage NFS mount is listed in Home > Infrastructure > Secondary
>>> Storage and the URL is correct.  Does this mean the secondary storage is
>>> unreachable?
>>>
>>>
>>> On Wed, May 7, 2014 at 10:26 AM, Ian Young <iyoung@ratespecial.com>wrote:
>>>
>>>> I reinstalled my single server CloudStack system yesterday, following
>>>> the quick start guide precisely.  The only difference was that I used
>>>> /var/primary and /var/secondary instead of /primary and /secondary, because
>>>> the /var partition on this machine is very large.  The UI installer reached
>>>> the point where it says "Creating system VMs (this may take a while)" but
>>>> never finished.  I left it overnight and it still hadn't completed.  This
>>>> is typically the step that fails, most of the times I've installed
>>>> CloudStack, so I imagine I must be making the same fundamental mistake each
>>>> time, and I'd like to know what that is.
>>>>
>>>> I checked management.log and it's in a loop where it creates a
>>>> secondary storage VM, fails to start it, destroys it, and tries again.  It
>>>> says Host 1 is unreachable but I'm using the correct password, SELinux is
>>>> permissive, and all the iptables rules are in place.  In what way is it
>>>> trying to connect to Host 1?  SSH?  NFS?  Here's a log excerpt of messages
>>>> related to the SSVM:
>>>>
>>>> http://pastebin.com/X11A51bh
>>>>
>>>> NFS appears to be functional, since CloudStack automatically mounted
>>>> the primary storage.
>>>>
>>>> Filesystem            Size  Used Avail Use% Mounted on
>>>> /dev/sda3              20G  1.8G   17G  10% /
>>>> tmpfs                  32G     0   32G   0% /dev/shm
>>>> /dev/sda1             194M   42M  143M  23% /boot
>>>> /dev/sda4             1.8T  1.9G  1.7T   1% /var
>>>> 192.168.100.6:/var/primary
>>>>                       1.8T  1.9G  1.7T   1%
>>>> /mnt/0594caa2-ceb4-36c6-9b13-0ff149a130af
>>>>
>>>> How can I identify whatever it is that's preventing the SSVM from
>>>> starting?  Here is another log excerpt, without any filtering:
>>>>
>>>> http://pastebin.com/XsPGJQik
>>>>
>>>
>>>
>>
>

--089e0115ff70c7609e04f93ba1ce--