Start scylla problem

1,101 views
Skip to first unread message

jianjun396@gmail.com

<jianjun396@gmail.com>
unread,
Jan 19, 2016, 12:44:43 AM1/19/16
to ScyllaDB users
I created a VM and installed centos 7.

Downloaded the latest 0.15 version, followed the instruction to start scylla, but doesn't work.

After 
systemctl start scylla-server

systemctl status scylla-server
shows it is active.

but cqlsh gives error.
Connection error: ('Unable to connect to any servers', {'127.0.0.1': error(111, "Tried connecting to [('127.0.0.1', 9042)]. Last error: Connection refused")})

also gives error
curl: (7) Failed connect to localhost:10000; Connection refused

If I run scylla manually, it hangs even for:
scylla --help-loggers

How can I find out what's wrong? 

Asias He

<asias@scylladb.com>
unread,
Jan 19, 2016, 1:17:24 AM1/19/16
to scylladb-users@googlegroups.com
Check the following:

What does  $ journalctl |grep scylla, say?

Is there any coredump in /var/lib/scylla/coredump?

Is scylla process still running, $ ps aux | grep scylla ?

Are you running on XFS?

--
You received this message because you are subscribed to the Google Groups "ScyllaDB users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.
Visit this group at https://groups.google.com/group/scylladb-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/d10dd920-19ea-4d69-a466-24b42604324e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--
Asias

jianjun396@gmail.com

<jianjun396@gmail.com>
unread,
Jan 19, 2016, 1:33:38 AM1/19/16
to ScyllaDB users
 journalctl |grep scylla

  Jan 18 22:29:44 localhost.localdomain sudo[33309]:     root : TTY=unknown ; PWD=/ ; USER=scylla ; COMMAND=/usr/bin/scylla --log-to-syslog 1 --log-to-stdout 0 --default-log-level info --developer-mode true --network-stack posix

No coredump.

ps aux | grep scylla

root      33309  0.0  0.0 196000  2816 ?        Ss   22:29   0:00 sudo -E -u scylla /usr/bin/scylla --log-to-syslog 1 --log-to-stdout 0 --default-log-level info --developer-mode true --network-stack posix
scylla    33315 99.6  0.3 17180007676 12688 ?   R    22:29   2:38 /usr/bin/scylla --log-to-syslog 1 --log-to-stdout 0 --default-log-level info --developer-mode true --network-stack posix

I'm not running on xfs, so I turned developer-mode on.

Avi Kivity

<avi@scylladb.com>
unread,
Jan 19, 2016, 2:39:12 AM1/19/16
to scylladb-users@googlegroups.com
On 01/19/2016 07:44 AM, jianj...@gmail.com wrote:
I created a VM and installed centos 7.


How much memory does your VM have?  Prior to version 0.16, Scylla has problems with less that 2GB of memory.


jianjun396@gmail.com

<jianjun396@gmail.com>
unread,
Jan 19, 2016, 11:40:24 PM1/19/16
to ScyllaDB users
The vm got 4G memory.

Avi Kivity

<avi@scylladb.com>
unread,
Jan 20, 2016, 3:29:20 AM1/20/16
to scylladb-users@googlegroups.com
What does 'ss -ltnp' show?

Joel Jacobson

<joel@joeljacobson.com>
unread,
Jan 21, 2016, 3:37:56 AM1/21/16
to ScyllaDB users, jianjun396@gmail.com
Looks like the service is running. Try passing the host: ``cqlsh -h $IP ``

I would also take a look at nodetool status. ``nodetool -h $iP status``

jianjun396@gmail.com

<jianjun396@gmail.com>
unread,
Jan 21, 2016, 11:50:34 PM1/21/16
to ScyllaDB users, jianjun396@gmail.com
I use localhost. cqlsh localhost or cqlsh 127.0.0.1.
None of it works.

ss -ltnp

State      Recv-Q Send-Q                         Local Address:Port                                        Peer Address:Port              
LISTEN     0      5                              192.168.122.1:53                                                     *:*                  
LISTEN     0      128                                        *:22                                                     *:*                  
LISTEN     0      128                                127.0.0.1:631                                                    *:*                  
LISTEN     0      100                                127.0.0.1:25                                                     *:*                  
LISTEN     0      128                                       :::22                                                    :::*                  
LISTEN     0      128                                      ::1:631                                                   :::*                  
LISTEN     0      100                                      ::1:25                                                    :::*   


I upgrade to 0.16 version. Now 
systemctl start scylla-server

hangs.
The virtual memory of scylla shows 16T.

Asias He

<asias@scylladb.com>
unread,
Jan 22, 2016, 1:24:39 AM1/22/16
to scylladb-users@googlegroups.com, jianjun396@gmail.com
This means scylla does not boots up successfully. It does not listen on the cql port. 

 
I upgrade to 0.16 version. Now 
systemctl start scylla-server

hangs.

What does 

journalctl |grep scylla

say with 0.16 ?


 
The virtual memory of scylla shows 16T.

This is fine. It is not the memory scylla really consumes, it is just the virtual memory. 
 

--
You received this message because you are subscribed to the Google Groups "ScyllaDB users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.
Visit this group at https://groups.google.com/group/scylladb-users.

For more options, visit https://groups.google.com/d/optout.



--
Asias

Jianjun Wen

<jianjun396@gmail.com>
unread,
Jan 22, 2016, 1:32:09 AM1/22/16
to Asias He, scylladb-users@googlegroups.com
journalctl |grep scylla

Jan 21 22:30:17 localhost.localdomain sudo[98640]:   scylla : TTY=unknown ; PWD=/ ; USER=root ; COMMAND=/usr/lib/scylla/scylla_prepare

--
-Jianjun Wen
Wancube.com - 3D photography

Asias He

<asias@scylladb.com>
unread,
Jan 22, 2016, 2:01:21 AM1/22/16
to Jianjun Wen, Takuya ASADA, scylladb-users@googlegroups.com
On Fri, Jan 22, 2016 at 2:32 PM, Jianjun Wen <jianj...@gmail.com> wrote:
journalctl |grep scylla

Jan 21 22:30:17 localhost.localdomain sudo[98640]:   scylla : TTY=unknown ; PWD=/ ; USER=root ; COMMAND=/usr/lib/scylla/scylla_prepare


This looks bogus. At least we are supposed to see /usr/bin/scylla was started to run and some early messages printed by scylla. Perhaps scylla_prepare never completes for some reason. 

Takuya?



--
Asias

Asias He

<asias@scylladb.com>
unread,
Jan 22, 2016, 2:03:08 AM1/22/16
to Jianjun Wen, Takuya ASADA, scylladb-users@googlegroups.com
On Fri, Jan 22, 2016 at 3:00 PM, Asias He <as...@scylladb.com> wrote:


On Fri, Jan 22, 2016 at 2:32 PM, Jianjun Wen <jianj...@gmail.com> wrote:
journalctl |grep scylla

Jan 21 22:30:17 localhost.localdomain sudo[98640]:   scylla : TTY=unknown ; PWD=/ ; USER=root ; COMMAND=/usr/lib/scylla/scylla_prepare


This looks bogus. At least we are supposed to see /usr/bin/scylla was started to run and some early messages printed by scylla. Perhaps scylla_prepare never completes for some reason. 

Takuya?
 

In scylla-server.service,

  0 [Unit]
  1 Description=Scylla Server
  2 After=network.target libvirtd.service

Are network.target and libvirtd.service started properly?



--
Asias

Jianjun Wen

<jianjun396@gmail.com>
unread,
Jan 22, 2016, 2:03:20 AM1/22/16
to Asias He, Takuya ASADA, scylladb-users@googlegroups.com

I manually run scylla_prepare, finished quickly

jianjun396@gmail.com

<jianjun396@gmail.com>
unread,
Jan 29, 2016, 12:22:49 AM1/29/16
to ScyllaDB users, asias@scylladb.com, syuu@scylladb.com, jianjun396@gmail.com
Found the problem:
 hwloc_topology_load hangs on my vm.

lstopo hangs also.

Is this an known issue to centos7?
I'm using the latest centos7 version.
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-users+unsubscribe@googlegroups.com.
To post to this group, send email to scylladb-users@googlegroups.com.



--
Asias



--
-Jianjun Wen
Wancube.com - 3D photography



--
Asias

Avi Kivity

<avi@scylladb.com>
unread,
Jan 30, 2016, 11:20:14 AM1/30/16
to scylladb-users@googlegroups.com, asias@scylladb.com, syuu@scylladb.com, jianjun396@gmail.com
Ah, good progress.

This is not a known problem, on centos or otherwise.

Can you try an 'strace lstopo' to see what might be broken?

What hypervisor are you using?
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.

Jianjun Wen

<jianjun396@gmail.com>
unread,
Jan 30, 2016, 12:14:02 PM1/30/16
to Avi Kivity, scylladb-users@googlegroups.com, Asias He, Takuya ASADA
Tried strace, nothing useful.
I'm using vmware.

To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.
Visit this group at https://groups.google.com/group/scylladb-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/c7e61490-cb74-42e3-ad82-22f937dd2ea5%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.



--
Asias



--
-Jianjun Wen
Wancube.com - 3D photography



--
Asias
--
You received this message because you are subscribed to the Google Groups "ScyllaDB users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.
Visit this group at https://groups.google.com/group/scylladb-users.

Pekka Enberg

<penberg@scylladb.com>
unread,
Feb 3, 2016, 4:31:41 AM2/3/16
to ScyllaDB users, Avi Kivity, Asias He, Takuya ASADA, Jianjun Wen
On Sat, Jan 30, 2016 at 7:14 PM, Jianjun Wen <jianj...@gmail.com> wrote:
> Tried strace, nothing useful.

Does setting the HWLOC_DEBUG_VERBOSE environment variable to 1 make
hwloc produce any debugging info?

- Pekka

Jianjun Wen

<jianjun396@gmail.com>
unread,
Feb 3, 2016, 12:15:59 PM2/3/16
to Pekka Enberg, Asias He, scylladb-users@googlegroups.com, Takuya ASADA, Avi Kivity

I worked with Holocaust team. They found it is VMware problem. Hwloc pushed an fix.  It works good for me.

Avi Kivity

<avi@scylladb.com>
unread,
Feb 5, 2016, 10:27:32 AM2/5/16
to Jianjun Wen, Pekka Enberg, Asias He, scylladb-users@googlegroups.com, Takuya ASADA
Cool, let's hope they release a new version soon and that it gets propagated into the various distributions.
Reply all
Reply to author
Forward
0 new messages