Connections from Linux client take a long time to
Justin reported that a simple dbaccess query took ~2.5s from one Linux client but only 0.04s from an AIX box, both over TCP. Respondents suspected the server's reverse hostname/DNS lookup on connect, suggesting /etc/hosts entries on both ends, limiting /etc/nsswitch.conf 'hosts:' to files, fixing DNS, and disabling IPv6 (IFX_DISABLE_IPV6, or BIND4 in nsswitch) since bad IPv6 resolution causes long waits. Justin found reverse lookups fast and /etc/hosts entries only cut the time to ~1.8s; he was about to try disabling IPv6 on the client. No resolution is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Hi,
I have a Linux server that seems to take a long time to make connections. On a
working (AIX) server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess rmca_test -
Database selected.
(expression)
2018-01-28 14:44:16.000
1 row(s) retrieved.
Database closed.
real 0m0.04s
user 0m0.02s
sys 0m0.01s
whereas on the Linux server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess sysmaster -
Database selected.
(expression)
2018-01-28 14:44:09.000
1 row(s) retrieved.
Database closed.
real 0m2.477s
user 0m0.012s
sys 0m0.002s
Any thoughts on where I should start looking?
Thanks,
-Justin
Is one using a shared memory connection and the other a network connection?
j.
> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
>
> Hi,
>
> I have a Linux server that seems to take a long time to make connections. On
a
> working (AIX) server, I get this:
>
> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess rmca_test
> -
>
> Database selected.
>
> (expression)
>
> 2018-01-28 14:44:16.000
>
> 1 row(s) retrieved.
>
> Database closed.
>
> real 0m0.04s
> user 0m0.02s
> sys 0m0.01s
>
> whereas on the Linux server, I get this:
>
> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess sysmaster
> -
>
> Database selected.
>
> (expression)
>
> 2018-01-28 14:44:09.000
>
> 1 row(s) retrieved.
>
> Database closed.
>
> real 0m2.477s
> user 0m0.012s
> sys 0m0.002s
>
> Any thoughts on where I should start looking?
>
> Thanks,
> -Justin
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
If it is on the network connection, try adding an entry in the /etc/hosts
file for the quest connection.
On 01/28/2018 05:42 PM, Jack Parker wrote:
> Is one using a shared memory connection and the other a network connection?
>
> j.
>
>> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
>>
>> Hi,
>>
>> I have a Linux server that seems to take a long time to make connections. On
> a
>> working (AIX) server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> rmca_test
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:16.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m0.04s
>> user 0m0.02s
>> sys 0m0.01s
>>
>> whereas on the Linux server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> sysmaster
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:09.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m2.477s
>> user 0m0.012s
>> sys 0m0.002s
>>
>> Any thoughts on where I should start looking?
>>
>> Thanks,
>> -Justin
>>
>>
>>
>
*******************************************************************************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
I tried that, but no change
> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
>
> If it is on the network connection, try adding an entry in the /etc/hosts
> file for the quest connection.
>
> On 01/28/2018 05:42 PM, Jack Parker wrote:
>> Is one using a shared memory connection and the other a network connection?
>>
>> j.
>>
>>> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
>> wrote:
>>>
>>> Hi,
>>>
>>> I have a Linux server that seems to take a long time to make connections.
> On
>> a
>>> working (AIX) server, I get this:
>>>
>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
>> rmca_test
>>> -
>>>
>>> Database selected.
>>>
>>> (expression)
>>>
>>> 2018-01-28 14:44:16.000
>>>
>>> 1 row(s) retrieved.
>>>
>>> Database closed.
>>>
>>> real 0m0.04s
>>> user 0m0.02s
>>> sys 0m0.01s
>>>
>>> whereas on the Linux server, I get this:
>>>
>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
>> sysmaster
>>> -
>>>
>>> Database selected.
>>>
>>> (expression)
>>>
>>> 2018-01-28 14:44:09.000
>>>
>>> 1 row(s) retrieved.
>>>
>>> Database closed.
>>>
>>> real 0m2.477s
>>> user 0m0.012s
>>> sys 0m0.002s
>>>
>>> Any thoughts on where I should start looking?
>>>
>>> Thanks,
>>> -Justin
>>>
>>>
>>>
>>
>
*******************************************************************************
>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>
>>
>>
>>
>
*******************************************************************************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
No, they are both network connections
> On Jan 28, 2018, at 3:42 PM, Jack Parker <jack.parker4@verizon.net> wrote:
>
> Is one using a shared memory connection and the other a network connection?
>
> j.
>
>> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
>>
>> Hi,
>>
>> I have a Linux server that seems to take a long time to make connections. On
> a
>> working (AIX) server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> rmca_test
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:16.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m0.04s
>> user 0m0.02s
>> sys 0m0.01s
>>
>> whereas on the Linux server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> sysmaster
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:09.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m2.477s
>> user 0m0.012s
>> sys 0m0.002s
>>
>> Any thoughts on where I should start looking?
>>
>> Thanks,
>> -Justin
>>
>>
>>
>
*******************************************************************************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
Did you do it at both ends
ie the server name/ip in the client hosts
and the client name/ip in the server hosts?
I suspect its the reverse dns lookup on the server
Clive
> On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
>
> I tried that, but no change
>
>> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
>>
>> If it is on the network connection, try adding an entry in the /etc/hosts
>> file for the quest connection.
>>
>> On 01/28/2018 05:42 PM, Jack Parker wrote:
>>> Is one using a shared memory connection and the other a network connection?
>>>
>>> j.
>>>
>>>> On Jan 28, 2018, at 6:36 PM, Justin Killen
> <jkillen@allamericanasphalt.com>
>>> wrote:
>>>>
>>>> Hi,
>>>>
>>>> I have a Linux server that seems to take a long time to make connections.
>> On
>>> a
>>>> working (AIX) server, I get this:
>>>>
>>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
>>> rmca_test
>>>> -
>>>>
>>>> Database selected.
>>>>
>>>> (expression)
>>>>
>>>> 2018-01-28 14:44:16.000
>>>>
>>>> 1 row(s) retrieved.
>>>>
>>>> Database closed.
>>>>
>>>> real 0m0.04s
>>>> user 0m0.02s
>>>> sys 0m0.01s
>>>>
>>>> whereas on the Linux server, I get this:
>>>>
>>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
>>> sysmaster
>>>> -
>>>>
>>>> Database selected.
>>>>
>>>> (expression)
>>>>
>>>> 2018-01-28 14:44:09.000
>>>>
>>>> 1 row(s) retrieved.
>>>>
>>>> Database closed.
>>>>
>>>> real 0m2.477s
>>>> user 0m0.012s
>>>> sys 0m0.002s
>>>>
>>>> Any thoughts on where I should start looking?
>>>>
>>>> Thanks,
>>>> -Justin
>>>>
>>>>
>>>>
>>>
>>
>
*******************************************************************************
>>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>>
>>>
>>>
>>>
>>
>
*******************************************************************************
>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>
>>
>>
>>
>
*******************************************************************************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
It's for sure an issue with reverse DNS.
You can try:
1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
the engine (if you don't plan to use IPV6)
3- Properly configure your DNS infra-structure
Regards.
On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
> Did you do it at both ends
>
> ie the server name/ip in the client hosts
> and the client name/ip in the server hosts?
>
> I suspect its the reverse dns lookup on the server
>
> Clive
>
> > On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
> >
> > I tried that, but no change
> >
> >> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
> >>
> >> If it is on the network connection, try adding an entry in the
> /etc/hosts
> >> file for the quest connection.
> >>
> >> On 01/28/2018 05:42 PM, Jack Parker wrote:
> >>> Is one using a shared memory connection and the other a network
> connection?
> >>>
> >>> j.
> >>>
> >>>> On Jan 28, 2018, at 6:36 PM, Justin Killen
> > <jkillen@allamericanasphalt.com>
> >>> wrote:
> >>>>
> >>>> Hi,
> >>>>
> >>>> I have a Linux server that seems to take a long time to make
> connections.
> >> On
> >>> a
> >>>> working (AIX) server, I get this:
> >>>>
> >>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> >>> rmca_test
> >>>> -
> >>>>
> >>>> Database selected.
> >>>>
> >>>> (expression)
> >>>>
> >>>> 2018-01-28 14:44:16.000
> >>>>
> >>>> 1 row(s) retrieved.
> >>>>
> >>>> Database closed.
> >>>>
> >>>> real 0m0.04s
> >>>> user 0m0.02s
> >>>> sys 0m0.01s
> >>>>
> >>>> whereas on the Linux server, I get this:
> >>>>
> >>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> >>> sysmaster
> >>>> -
> >>>>
> >>>> Database selected.
> >>>>
> >>>> (expression)
> >>>>
> >>>> 2018-01-28 14:44:09.000
> >>>>
> >>>> 1 row(s) retrieved.
> >>>>
> >>>> Database closed.
> >>>>
> >>>> real 0m2.477s
> >>>> user 0m0.012s
> >>>> sys 0m0.002s
> >>>>
> >>>> Any thoughts on where I should start looking?
> >>>>
> >>>> Thanks,
> >>>> -Justin
> >>>>
> >>>>
> >>>>
> >>>
> >>
> >
> ************************************************************
> *******************
> >>>> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>>>
> >>>
> >>>
> >>>
> >>
> >
> ************************************************************
> *******************
> >>> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>>
> >>
> >>
> >>
> >
> ************************************************************
> *******************
> >> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>
> >
> >
> >
> ************************************************************
> *******************
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
>
>
> ************************************************************
> *******************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
Just in case this is not clear:
Upon receiving a connect request, the Informix server does a lookup of the
cient hostname to verify, that this hostname is known for the given IP
address. This is what is referred to as "reverse hostname/DNS lookup".
If this happens via DNS and there is a problem with this lookup (e.g. not
satisfied by the first DNS server), then more time may be spent (waiting for
the answer from the first DNS server, then possibly trying and waiting for
subsequent DNS server(s)). This can cause quite a substantial delay.
Putting all the required info into local files (/etc/hosts) on both machines
(client and server) and configuring things accordingly (usually in
/etc/resolv.conf) can considerably shorten the lookup times by avoiding th
euse of DNS on both sides. This probably is not the final solution, but it
serves to verify whether DNS lookup is the cause of the delay. Once verified
and DNS lookup configured properly, the entries in local files shouldn't be
needed anymore.
TIA, Martin
--
Martin Fuerderer
Informix Development Germany
e-mail: martin.furderer@hcl.com
HCL Technologies Germany GmbH
Frankfurter Ring 17
80807 Munich, Germany
http://www.hcltech.com/de
--DISCLAIMER--
--------------------------------------------------------------------------------
----------------------------
This document is intended for transmission to the named recipient only. If you
are not that person, you should note that legal rights reside in this document
and you are not authorized to access, read, disclose, copy, use or otherwise
deal with it and any such actions are prohibited and may be unlawful. The
views expressed in this document are not necessarily those of HCL Technologies
Ltd. Notice is hereby given that no representation, contract or other binding
obligation shall be created by this e-mail, which must be interpreted
accordingly. Any representations, contractual rights or obligations shall be
separately communicated in writing and signed in the original by a duly
authorized officer of the relevant company.
--------------------------------------------------------------------------------
----------------------------
________________________________
From: ids-bounces@iiug.org <ids-bounces@iiug.org> on behalf of Fernando Nunes
<domusonline@gmail.com>
Sent: Monday, January 29, 2018 15:33
To: ids@iiug.org
Subject: Re: Connections from Linux client take a long .... [40594]
It's for sure an issue with reverse DNS.
You can try:
1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
the engine (if you don't plan to use IPV6)
3- Properly configure your DNS infra-structure
Regards.
On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
> Did you do it at both ends
>
> ie the server name/ip in the client hosts
> and the client name/ip in the server hosts?
>
> I suspect its the reverse dns lookup on the server
>
> Clive
>
> > On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
> >
> > I tried that, but no change
> >
> >> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
> >>
> >> If it is on the network connection, try adding an entry in the
> /etc/hosts
> >> file for the quest connection.
> >>
> >> On 01/28/2018 05:42 PM, Jack Parker wrote:
> >>> Is one using a shared memory connection and the other a network
> connection?
> >>>
> >>> j.
> >>>
> >>>> On Jan 28, 2018, at 6:36 PM, Justin Killen
> > <jkillen@allamericanasphalt.com>
> >>> wrote:
> >>>>
> >>>> Hi,
> >>>>
> >>>> I have a Linux server that seems to take a long time to make
> connections.
> >> On
> >>> a
> >>>> working (AIX) server, I get this:
> >>>>
> >>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> >>> rmca_test
> >>>> -
> >>>>
> >>>> Database selected.
> >>>>
> >>>> (expression)
> >>>>
> >>>> 2018-01-28 14:44:16.000
> >>>>
> >>>> 1 row(s) retrieved.
> >>>>
> >>>> Database closed.
> >>>>
> >>>> real 0m0.04s
> >>>> user 0m0.02s
> >>>> sys 0m0.01s
> >>>>
> >>>> whereas on the Linux server, I get this:
> >>>>
> >>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> >>> sysmaster
> >>>> -
> >>>>
> >>>> Database selected.
> >>>>
> >>>> (expression)
> >>>>
> >>>> 2018-01-28 14:44:09.000
> >>>>
> >>>> 1 row(s) retrieved.
> >>>>
> >>>> Database closed.
> >>>>
> >>>> real 0m2.477s
> >>>> user 0m0.012s
> >>>> sys 0m0.002s
> >>>>
> >>>> Any thoughts on where I should start looking?
> >>>>
> >>>> Thanks,
> >>>> -Justin
> >>>>
> >>>>
> >>>>
> >>>
> >>
> >
> ************************************************************
> *******************
> >>>> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>>>
> >>>
> >>>
> >>>
> >>
> >
> ************************************************************
> *******************
> >>> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>>
> >>
> >>
> >>
> >
> ************************************************************
> *******************
> >> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>
> >
> >
> >
> ************************************************************
> *******************
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
>
>
> ************************************************************
> *******************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
https://apac01.safelinks.protection.outlook.com/?url=http%3A%2F%2Finformix-techn
ology.blogspot.com&data=02%7C01%7Cmartin.furderer%40hcl.com%7C72c78e15062c48f3d3
2808d5672568f5%7C189de737c93a4f5a8b686f4ca9941912%7C0%7C0%7C636528332766568533&s
data=8qoI9oaYcD1pzuMrXH5Q%2F7hq1kgcSAJth6AJy0Gs2%2Bs%3D&reserved=0
My email works... but I don't check it frequently...
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
::DISCLAIMER::
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------
--------------------------------------
The contents of this e-mail and any attachment(s) are confidential and
intended for the named recipient(s) only. E-mail transmission is not
guaranteed to be secure or error-free as information could be intercepted,
corrupted, lost, destroyed, arrive late or incomplete, or may contain viruses
in transmission. The e mail and its contents (with or without referred errors)
shall therefore not attach any liability on the ori
Look at dns and IPv6 setting. Might turn off ipv6.
Sent from my iPhone
> On Jan 28, 2018, at 3:42 PM, Jack Parker <jack.parker4@verizon.net> wrote:
>
> Is one using a shared memory connection and the other a network connection?
>
> j.
>
>> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
>>
>> Hi,
>>
>> I have a Linux server that seems to take a long time to make connections. On
> a
>> working (AIX) server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> rmca_test
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:16.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m0.04s
>> user 0m0.02s
>> sys 0m0.01s
>>
>> whereas on the Linux server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> sysmaster
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:09.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m2.477s
>> user 0m0.012s
>> sys 0m0.002s
>>
>> Any thoughts on where I should start looking?
>>
>> Thanks,
>> -Justin
>>
>>
>>
>
*******************************************************************************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
Ran into this as well on AIX and Linux.
The environment variable worked like a charm but I view it as a work around
for poor configuration.
In the /etc/nsswitch.conf if there is a BIND or BIND6 option on the hosts line
it will try to resolve using IPv6 if available. Changing to BIND4 forces it to
only resolve using IPv4. Better than maintaining host files on each system
which is what happen if you disable BIND all together. Adding to the host file
with BIND set did not help in my experience as it waited for response from all
verse using the first available (might be able to configure this but found it
unnecessary).
What causes the issue? If the DNS system is not setup properly for the IPv6 or
the system should not have IPv6 enabled this could be the source.
Examples:
In my one of my cases the system should not have IPv6 set and removing it from
the nic fix the problem as there was not IPv6 network to check (Linux).
In another the system admin wanted IPv6 but our DNS is not setup to allow
resolution so we changed to BIND4 (AIX).
Sadly I dont have a setup where IPv6 resolves to anything but look forward to
the day.
Hope this helps,
Eric Rowell
> On Jan 29, 2018, at 09:33, Fernando Nunes <domusonline@gmail.com> wrote:
>
> It's for sure an issue with reverse DNS.
> You can try:
> 1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
> 2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
> the engine (if you don't plan to use IPV6)
> 3- Properly configure your DNS infra-structure
>
> Regards.
>
>> On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
>>
>> Did you do it at both ends
>>
>> ie the server name/ip in the client hosts
>> and the client name/ip in the server hosts?
>>
>> I suspect its the reverse dns lookup on the server
>>
>> Clive
>>
>>> On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
>> wrote:
>>>
>>> I tried that, but no change
>>>
>>>> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
>>>>
>>>> If it is on the network connection, try adding an entry in the
>> /etc/hosts
>>>> file for the quest connection.
>>>>
>>>>> On 01/28/2018 05:42 PM, Jack Parker wrote:
>>>>> Is one using a shared memory connection and the other a network
>> connection?
>>>>>
>>>>> j.
>>>>>
>>>>>> On Jan 28, 2018, at 6:36 PM, Justin Killen
>>> <jkillen@allamericanasphalt.com>
>>>>> wrote:
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> I have a Linux server that seems to take a long time to make
>> connections.
>>>> On
>>>>> a
>>>>>> working (AIX) server, I get this:
>>>>>>
>>>>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
>>>>> rmca_test
>>>>>> -
>>>>>>
>>>>>> Database selected.
>>>>>>
>>>>>> (expression)
>>>>>>
>>>>>> 2018-01-28 14:44:16.000
>>>>>>
>>>>>> 1 row(s) retrieved.
>>>>>>
>>>>>> Database closed.
>>>>>>
>>>>>> real 0m0.04s
>>>>>> user 0m0.02s
>>>>>> sys 0m0.01s
>>>>>>
>>>>>> whereas on the Linux server, I get this:
>>>>>>
>>>>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
>>>>> sysmaster
>>>>>> -
>>>>>>
>>>>>> Database selected.
>>>>>>
>>>>>> (expression)
>>>>>>
>>>>>> 2018-01-28 14:44:09.000
>>>>>>
>>>>>> 1 row(s) retrieved.
>>>>>>
>>>>>> Database closed.
>>>>>>
>>>>>> real 0m2.477s
>>>>>> user 0m0.012s
>>>>>> sys 0m0.002s
>>>>>>
>>>>>> Any thoughts on where I should start looking?
>>>>>>
>>>>>> Thanks,
>>>>>> -Justin
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>> ************************************************************
>> *******************
>>>>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>> ************************************************************
>> *******************
>>>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>>>
>>>>
>>>>
>>>>
>>>
>> ************************************************************
>> *******************
>>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>>
>>>
>>>
>>>
>> ************************************************************
>> *******************
>>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>>
>>
>>
>> ************************************************************
>> *******************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>>
>
> --
> Fernando Nunes
> Portugal
>
> http://informix-technology.blogspot.com
> My email works... but I don't check it frequently...
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
Reverse DNS lookup time is 11ms - I don't think this is the problem.
Again, I have other clients that are connecting to these instances without any
issue, it's just this specific application server that is having issues.
-Justin
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Fernando
Nunes
Sent: Monday, January 29, 2018 6:34 AM
To: ids@iiug.org
Subject: Re: Connections from Linux client take a long .... [40594]
It's for sure an issue with reverse DNS.
You can try:
1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
the engine (if you don't plan to use IPV6)
3- Properly configure your DNS infra-structure
Regards.
On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
> Did you do it at both ends
>
> ie the server name/ip in the client hosts and the client name/ip in
> the server hosts?
>
> I suspect its the reverse dns lookup on the server
>
> Clive
>
> > On 29 Jan 2018, at 04:18, Justin Killen
> > <jkillen@allamericanasphalt.com>
> wrote:
> >
> > I tried that, but no change
> >
> >> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
> >>
> >> If it is on the network connection, try adding an entry in the
> /etc/hosts
> >> file for the quest connection.
> >>
> >> On 01/28/2018 05:42 PM, Jack Parker wrote:
> >>> Is one using a shared memory connection and the other a network
> connection?
> >>>
> >>> j.
> >>>
> >>>> On Jan 28, 2018, at 6:36 PM, Justin Killen
> > <jkillen@allamericanasphalt.com>
> >>> wrote:
> >>>>
> >>>> Hi,
> >>>>
> >>>> I have a Linux server that seems to take a long time to make
> connections.
> >> On
> >>> a
> >>>> working (AIX) server, I get this:
> >>>>
> >>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" |
> >>>> dbaccess
> >>> rmca_test
> >>>> -> >>>>
> >>>> Database selected.
> >>>>
> >>>> (expression)
> >>>>
> >>>> 2018-01-28 14:44:16.000
> >>>>
> >>>> 1 row(s) retrieved.
> >>>>
> >>>> Database closed.
> >>>>
> >>>> real 0m0.04s
> >>>> user 0m0.02s
> >>>> sys 0m0.01s
> >>>>
> >>>> whereas on the Linux server, I get this:
> >>>>
> >>>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" |
> >>>> dbaccess> >>> sysmaster
> >>>> -
> >>>>
> >>>> Database selected.
> >>>>
> >>>> (expression)
> >>>>
> >>>> 2018-01-28 14:44:09.000
> >>>>
> >>>> 1 row(s) retrieved.
> >>>>
> >>>> Database closed.
> >>>>
> >>>> real 0m2.477s
> >>>> user 0m0.012s
> >>>> sys 0m0.002s
> >>>>
> >>>> Any thoughts on where I should start looking?
> >>>>
> >>>> Thanks,
> >>>> -Justin
> >>>>
> >>>>
> >>>>
> >>>
> >>
> >
> ************************************************************
> *******************
> >>>> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>>>
> >>>
> >>>
> >>>
> >>
> >
> ************************************************************
> *******************
> >>> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>>
> >>
> >>
> >>
> >
> ************************************************************
> *******************
> >> Forum Note: Use "Reply" to post a response in the discussion forum.
> >>
> >
> >
> >
> ************************************************************
> *******************
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
>
>
> ************************************************************
> *******************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
I added the entries to /etc/hosts, the sample query is now taking around 1.8
seconds instead of 2 - I still don't think DNS is the issue.
The server has IPv6 disabled, but it looks like the client doesn't - I'll give
that a try.
-Justin
On Jan 31, 2018, at 11:01 PM, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
Reverse DNS lookup time is 11ms - I don't think this is the problem.
Again, I have other clients that are connecting to these instances without any
issue, it's just this specific application server that is having issues.
-Justin
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Fernando
Nunes
Sent: Monday, January 29, 2018 6:34 AM
To: ids@iiug.org
Subject: Re: Connections from Linux client take a long .... [40594]
It's for sure an issue with reverse DNS.
You can try:
1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
the engine (if you don't plan to use IPV6)
3- Properly configure your DNS infra-structure
Regards.
On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
Did you do it at both ends
ie the server name/ip in the client hosts and the client name/ip in
the server hosts?
I suspect its the reverse dns lookup on the server
Clive
On 29 Jan 2018, at 04:18, Justin Killen
<jkillen@allamericanasphalt.com>
wrote:
I tried that, but no change
On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
If it is on the network connection, try adding an entry in the
/etc/hosts
file for the quest connection.
On 01/28/2018 05:42 PM, Jack Parker wrote:
Is one using a shared memory connection and the other a network
connection?
j.
On Jan 28, 2018, at 6:36 PM, Justin Killen
<jkillen@allamericanasphalt.com>
wrote:
Hi,
I have a Linux server that seems to take a long time to make
connections.
On
a
working (AIX) server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" |
dbaccess
rmca_test
-
Database selected.
(expression)
2018-01-28 14:44:16.000
1 row(s) retrieved.
Database closed.
real 0m0.04s
user 0m0.02s
sys 0m0.01s
whereas on the Linux server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" |
dbaccesssysmaster
-
Database selected.
(expression)
2018-01-28 14:44:09.000
1 row(s) retrieved.
Database closed.
real 0m2.477s
user 0m0.012s
sys 0m0.002s
Any thoughts on where I should start looking?
Thanks,
-Justin
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
You can test 'anything connection', incl. name resolution, tcp connection,
authentication, etc. by simply doing
for i in {1..100};do echo "" | dbaccess mydatabase@myserver ; done
This would simply connect, establish a session and disconnect a hundred
times in a row and you'd exclude any SQL activity eating time (provided no
sysdbopen UDR).
If this is fast, you'd know your query is the time consumer, else you can
concentrate on connection details.
From: "Justin Killen" <jkillen@allamericanasphalt.com>
To: ids@iiug.org
Date: 02/01/2018 08:17 AM
Subject: Re: Connections from Linux client take a long .... [40618]
Sent by: ids-bounces@iiug.org
I added the entries to /etc/hosts, the sample query is now taking around
1.8
seconds instead of 2 - I still don't think DNS is the issue.
The server has IPv6 disabled, but it looks like the client doesn't - I'll
give
that a try.
-Justin
On Jan 31, 2018, at 11:01 PM, Justin Killen
<jkillen@allamericanasphalt.com>
wrote:
Reverse DNS lookup time is 11ms - I don't think this is the problem.
Again, I have other clients that are connecting to these instances without
any
issue, it's just this specific application server that is having issues.
-Justin
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Fernando
Nunes
Sent: Monday, January 29, 2018 6:34 AM
To: ids@iiug.org
Subject: Re: Connections from Linux client take a long .... [40594]
It's for sure an issue with reverse DNS.
You can try:
1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
the engine (if you don't plan to use IPV6)
3- Properly configure your DNS infra-structure
Regards.
On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
Did you do it at both ends
ie the server name/ip in the client hosts and the client name/ip in
the server hosts?
I suspect its the reverse dns lookup on the server
Clive
On 29 Jan 2018, at 04:18, Justin Killen
<jkillen@allamericanasphalt.com>
wrote:
I tried that, but no change
On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
If it is on the network connection, try adding an entry in the
/etc/hosts
file for the quest connection.
On 01/28/2018 05:42 PM, Jack Parker wrote:
Is one using a shared memory connection and the other a network
connection?
j.
On Jan 28, 2018, at 6:36 PM, Justin Killen
<jkillen@allamericanasphalt.com>
wrote:
Hi,
I have a Linux server that seems to take a long time to make
connections.
On
a
working (AIX) server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" |
dbaccess
rmca_test
-
Database selected.
(expression)
2018-01-28 14:44:16.000
1 row(s) retrieved.
Database closed.
real 0m0.04s
user 0m0.02s
sys 0m0.01s
whereas on the Linux server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" |
dbaccesssysmaster
-
Database selected.
(expression)
2018-01-28 14:44:09.000
1 row(s) retrieved.
Database closed.
real 0m2.477s
user 0m0.012s
sys 0m0.002s
Any thoughts on where I should start looking?
Thanks,
-Justin
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
I added the environment variable on both the client and the server per
https://www.ibm.com/support/knowledgecenter/en/SSGU8G_11.50.0/com.ibm.admin.doc/
ids_admin_0198.htm
Then I restarted the cluster instances and connection managers - No change in
behavior. I also added the empty file $INFORMIXDIR/etc/IFX_DISABLE_IPV6 and
restarted - still no change in behavior.
All 4 hosts (the client, the primary, the secondary, and an RSS) have their
static IPs in their hosts files, and nsswitch.conf is set to only files.
All 4 hosts also have IPv6 disabled via:
# cat /etc/sysctl.d/disableipv6.conf
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
net.ipv6.conf.lo.disable_ipv6 = 1
The Informix servers are RHEL 7.4 on bare hardware, the client is CentOS 6.9
running as a hyper-v VM.
Per another suggestion, I changed my test to the following to separate out any
query lag:
time echo "" | dbaccess rmca_test
The average run time of 5 runs was 1.561 seconds, which is way too high.
I tried to curl just to confirm timings, and I get this:
> curl -vw "@curl-format.txt" 192.168.1.42:10022 -o /dev/null -s
* About to connect() to 192.168.1.42 port 10022 (#0)
* Trying 192.168.1.42... connected
* Connected to 192.168.1.42 (192.168.1.42) port 10022 (#0)
> GET / HTTP/1.1
> User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.27.1
zlib/1.2.3 libidn/1.18 libssh2/1.4.2
> Host: 192.168.1.42:10022
> Accept: */*
>
* Empty reply from server
* Connection #0 to host 192.168.1.42 left intact
* Closing connection #0
n time_namelookup: 0.000
n time_connect: 0.015
n time_appconnect: 0.000
n time_pretransfer: 0.015
n time_redirect: 0.000
n time_starttransfer: 0.061
n ----------
n time_total: 0.061
So, that pretty much rules out basic connectivity. I also grabbed a pcap if
that's useful:
https://www.dropbox.com/s/1si2t1zswk5ht0m/informix_empty_connect.pcap?dl=0
Any other suggestions?
-Justin
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Justin
Killen
Sent: Wednesday, January 31, 2018 11:17 PM
To: ids@iiug.org
Subject: Re: Connections from Linux client take a long .... [40618]
I added the entries to /etc/hosts, the sample query is now taking around 1.8
seconds instead of 2 - I still don't think DNS is the issue.
The server has IPv6 disabled, but it looks like the client doesn't - I'll give
that a try.
-Justin
On Jan 31, 2018, at 11:01 PM, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
Reverse DNS lookup time is 11ms - I don't think this is the problem.
Again, I have other clients that are connecting to these instances without any
issue, it's just this specific application server that is having issues.
-Justin
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Fernando
Nunes
Sent: Monday, January 29, 2018 6:34 AM
To: ids@iiug.org
Subject: Re: Connections from Linux client take a long .... [40594]
It's for sure an issue with reverse DNS.
You can try:
1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
the engine (if you don't plan to use IPV6)
3- Properly configure your DNS infra-structure
Regards.
On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
Did you do it at both ends
ie the server name/ip in the client hosts and the client name/ip in the server
hosts?
I suspect its the reverse dns lookup on the server
Clive
On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
I tried that, but no change
On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
If it is on the network connection, try adding an entry in the /etc/hosts
file for the quest connection.
On 01/28/2018 05:42 PM, Jack Parker wrote:
Is one using a shared memory connection and the other a network connection?
j.
On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
Hi,
I have a Linux server that seems to take a long time to make connections.
On
a
working (AIX) server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess rmca_test
-
Database selected.
(expression)
2018-01-28 14:44:16.000
1 row(s) retrieved.
Database closed.
real 0m0.04s
user 0m0.02s
sys 0m0.01s
whereas on the Linux server, I get this:
time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess sysmaster
-
Database selected.
(expression)
2018-01-28 14:44:09.000
1 row(s) retrieved.
Database closed.
real 0m2.477s
user 0m0.012s
sys 0m0.002s
Any thoughts on where I should start looking?
Thanks,
-Justin
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
************************************************************
*******************
Forum Note: Use "Reply" to post a response in the discussion forum.
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Nor sure I'm reading this tcp trace right, but I'd say this this is an exchange of about 12-14 packages in either way and it appears there's a 150-200 ms delay for almost every reply from server to client while client would react pretty much instantaneously. So no specific server operation causing the majority of the overall 1.9 seconds duration obviously. What I cannot tell is whether this delay occurs within the Informix server or rather somewhere below it and within the network. Can we assume this trace has been collected on the client machine? What would it look like when capturing on server side? What's the general load on the Informix server, what on its host? E.g. wrt. cpu consumption?
In the hosts file to you have the bare hostnames or the fqdn?
If one or the other try with both
Clive
> On 3 Feb 2018, at 04:47, Justin Killen <jkillen@allamericanasphalt.com>
wrote:
>
> I added the environment variable on both the client and the server per
>
https://www.ibm.com/support/knowledgecenter/en/SSGU8G_11.50.0/com.ibm.admin.doc/
ids_admin_0198.htm
>
> Then I restarted the cluster instances and connection managers - No change in
> behavior. I also added the empty file $INFORMIXDIR/etc/IFX_DISABLE_IPV6 and
> restarted - still no change in behavior.
>
> All 4 hosts (the client, the primary, the secondary, and an RSS) have their
> static IPs in their hosts files, and nsswitch.conf is set to only files.
>
> All 4 hosts also have IPv6 disabled via:
> # cat /etc/sysctl.d/disableipv6.conf
> net.ipv6.conf.all.disable_ipv6 = 1
> net.ipv6.conf.default.disable_ipv6 = 1
> net.ipv6.conf.lo.disable_ipv6 = 1>
> The Informix servers are RHEL 7.4 on bare hardware, the client is CentOS 6.9
> running as a hyper-v VM.
>
> Per another suggestion, I changed my test to the following to separate out
any
> query lag:
> time echo "" | dbaccess rmca_test
>
> The average run time of 5 runs was 1.561 seconds, which is way too high.
>
> I tried to curl just to confirm timings, and I get this:
>
>> curl -vw "@curl-format.txt" 192.168.1.42:10022 -o /dev/null -s
> * About to connect() to 192.168.1.42 port 10022 (#0)
> * Trying 192.168.1.42... connected
> * Connected to 192.168.1.42 (192.168.1.42) port 10022 (#0)
>> GET / HTTP/1.1
>> User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.27.1
> zlib/1.2.3 libidn/1.18 libssh2/1.4.2
>> Host: 192.168.1.42:10022
>> Accept: */*
>>
> * Empty reply from server
> * Connection #0 to host 192.168.1.42 left intact
> * Closing connection #0
> n time_namelookup: 0.000
> n time_connect: 0.015
> n time_appconnect: 0.000
> n time_pretransfer: 0.015
> n time_redirect: 0.000
> n time_starttransfer: 0.061
> n ----------
> n time_total: 0.061
>
> So, that pretty much rules out basic connectivity. I also grabbed a pcap if
> that's useful:
> https://www.dropbox.com/s/1si2t1zswk5ht0m/informix_empty_connect.pcap?dl=0
>
> Any other suggestions?
>
> -Justin
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Justin
> Killen
> Sent: Wednesday, January 31, 2018 11:17 PM
> To: ids@iiug.org
> Subject: Re: Connections from Linux client take a long .... [40618]
>
> I added the entries to /etc/hosts, the sample query is now taking around 1.8
> seconds instead of 2 - I still don't think DNS is the issue.
>
> The server has IPv6 disabled, but it looks like the client doesn't - I'll
give
> that a try.
>
> -Justin
>
> On Jan 31, 2018, at 11:01 PM, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
>
> Reverse DNS lookup time is 11ms - I don't think this is the problem.
>
> Again, I have other clients that are connecting to these instances without
any
> issue, it's just this specific application server that is having issues.
>
> -Justin
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Fernando
> Nunes
> Sent: Monday, January 29, 2018 6:34 AM
> To: ids@iiug.org
> Subject: Re: Connections from Linux client take a long .... [40594]
>
> It's for sure an issue with reverse DNS.
> You can try:
> 1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
> 2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
> the engine (if you don't plan to use IPV6)
> 3- Properly configure your DNS infra-structure
>
> Regards.
>
> On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
>
> Did you do it at both ends
>
> ie the server name/ip in the client hosts and the client name/ip in the
server
> hosts?
>
> I suspect its the reverse dns lookup on the server
>
> Clive
>
> On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
>
> I tried that, but no change
>
> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
>
> If it is on the network connection, try adding an entry in the /etc/hosts
>
> file for the quest connection.
>
> On 01/28/2018 05:42 PM, Jack Parker wrote:
>
> Is one using a shared memory connection and the other a network connection?
>
> j.
>
> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
>
> wrote:
>
> Hi,
>
> I have a Linux server that seems to take a long time to make connections.
>
> On
>
> a
>
> working (AIX) server, I get this:
>
> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess rmca_test
>
> -
>
> Database selected.
>
> (expression)
>
> 2018-01-28 14:44:16.000
>
> 1 row(s) retrieved.
>
> Database closed.
>
> real 0m0.04s
> user 0m0.02s
> sys 0m0.01s
>
> whereas on the Linux server, I get this:
>
> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess sysmaster
>
> -
>
> Database selected.
>
> (expression)
>
> 2018-01-28 14:44:09.000
>
> 1 row(s) retrieved.
>
> Database closed.
>
> real 0m2.477s
> user 0m0.012s
> sys 0m0.002s
>
> Any thoughts on where I should start looking?
>
> Thanks,
> -Justin
>
> ************************************************************
> *******************
>
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> ************************************************************
> *******************
>
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> ************************************************************
> *******************
>
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> ************************************************************
> *******************
>
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> ************************************************************
> *******************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> --
> Fernando Nunes
> Portugal
>
> http://informix-technology.blogspot.com
> My email works... but I don't check it frequently...
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the
Yes, both short name and FQDN are specified.
> On Feb 4, 2018, at 12:29 AM, Clive Eisen <clive@serendipita.com> wrote:
>
> In the hosts file to you have the bare hostnames or the fqdn?
>
> If one or the other try with both
>
> Clive
>
>> On 3 Feb 2018, at 04:47, Justin Killen <jkillen@allamericanasphalt.com>
> wrote:
>>
>> I added the environment variable on both the client and the server per
>>
>
https://www.ibm.com/support/knowledgecenter/en/SSGU8G_11.50.0/com.ibm.admin.doc/
ids_admin_0198.htm
>>
>> Then I restarted the cluster instances and connection managers - No change
> in
>> behavior. I also added the empty file $INFORMIXDIR/etc/IFX_DISABLE_IPV6 and
>> restarted - still no change in behavior.
>>
>> All 4 hosts (the client, the primary, the secondary, and an RSS) have their
>> static IPs in their hosts files, and nsswitch.conf is set to only files.
>>
>> All 4 hosts also have IPv6 disabled via:
>> # cat /etc/sysctl.d/disableipv6.conf
>> net.ipv6.conf.all.disable_ipv6 = 1
>> net.ipv6.conf.default.disable_ipv6 = 1
>> net.ipv6.conf.lo.disable_ipv6 = 1>>
>> The Informix servers are RHEL 7.4 on bare hardware, the client is CentOS 6.9
>> running as a hyper-v VM.
>>
>> Per another suggestion, I changed my test to the following to separate out
> any
>> query lag:
>> time echo "" | dbaccess rmca_test
>>
>> The average run time of 5 runs was 1.561 seconds, which is way too high.
>>
>> I tried to curl just to confirm timings, and I get this:
>>
>>> curl -vw "@curl-format.txt" 192.168.1.42:10022 -o /dev/null -s
>> * About to connect() to 192.168.1.42 port 10022 (#0)
>> * Trying 192.168.1.42... connected
>> * Connected to 192.168.1.42 (192.168.1.42) port 10022 (#0)
>>> GET / HTTP/1.1
>>> User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.27.1
>> zlib/1.2.3 libidn/1.18 libssh2/1.4.2
>>> Host: 192.168.1.42:10022
>>> Accept: */*
>>>
>> * Empty reply from server
>> * Connection #0 to host 192.168.1.42 left intact
>> * Closing connection #0
>> n time_namelookup: 0.000
>> n time_connect: 0.015
>> n time_appconnect: 0.000
>> n time_pretransfer: 0.015
>> n time_redirect: 0.000
>> n time_starttransfer: 0.061
>> n ----------
>> n time_total: 0.061
>>
>> So, that pretty much rules out basic connectivity. I also grabbed a pcap if
>> that's useful:
>> https://www.dropbox.com/s/1si2t1zswk5ht0m/informix_empty_connect.pcap?dl=0
>>
>> Any other suggestions?
>>
>> -Justin
>>
>> -----Original Message-----
>> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Justin
>> Killen
>> Sent: Wednesday, January 31, 2018 11:17 PM
>> To: ids@iiug.org
>> Subject: Re: Connections from Linux client take a long .... [40618]
>>
>> I added the entries to /etc/hosts, the sample query is now taking around 1.8
>> seconds instead of 2 - I still don't think DNS is the issue.
>>
>> The server has IPv6 disabled, but it looks like the client doesn't - I'll
> give
>> that a try.
>>
>> -Justin
>>
>> On Jan 31, 2018, at 11:01 PM, Justin Killen <jkillen@allamericanasphalt.com>
>> wrote:
>>
>> Reverse DNS lookup time is 11ms - I don't think this is the problem.
>>
>> Again, I have other clients that are connecting to these instances without
> any
>> issue, it's just this specific application server that is having issues.
>>
>> -Justin
>>
>> -----Original Message-----
>> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> Fernando
>> Nunes
>> Sent: Monday, January 29, 2018 6:34 AM
>> To: ids@iiug.org
>> Subject: Re: Connections from Linux client take a long .... [40594]
>>
>> It's for sure an issue with reverse DNS.
>> You can try:
>> 1- Edit /etc/nsswitch.conf and on the "hosts:" entry leave only "files"
>> 2- Set IFX_DISABLE_IPV6 environment variable on the server side and restart
>> the engine (if you don't plan to use IPV6)
>> 3- Properly configure your DNS infra-structure
>>
>> Regards.
>>
>> On Mon, Jan 29, 2018 at 9:11 AM, Clive Eisen <clive@serendipita.com> wrote:
>>
>> Did you do it at both ends
>>
>> ie the server name/ip in the client hosts and the client name/ip in the
> server
>> hosts?
>>
>> I suspect its the reverse dns lookup on the server
>>
>> Clive
>>
>> On 29 Jan 2018, at 04:18, Justin Killen <jkillen@allamericanasphalt.com>
>> wrote:
>>
>> I tried that, but no change
>>
>> On Jan 28, 2018, at 4:00 PM, Ben Duncan <linux4ms@aim.com> wrote:
>>
>> If it is on the network connection, try adding an entry in the /etc/hosts
>>
>> file for the quest connection.
>>
>> On 01/28/2018 05:42 PM, Jack Parker wrote:
>>
>> Is one using a shared memory connection and the other a network connection?
>>
>> j.
>>
>> On Jan 28, 2018, at 6:36 PM, Justin Killen <jkillen@allamericanasphalt.com>
>>
>> wrote:
>>
>> Hi,
>>
>> I have a Linux server that seems to take a long time to make connections.
>>
>> On
>>
>> a
>>
>> working (AIX) server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> rmca_test
>>
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:16.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m0.04s
>> user 0m0.02s
>> sys 0m0.01s
>>
>> whereas on the Linux server, I get this:
>>
>> time echo "SELECT CURRENT FROM systables WHERE tabid=1;" | dbaccess
> sysmaster
>>
>> -
>>
>> Database selected.
>>
>> (expression)
>>
>> 2018-01-28 14:44:09.000
>>
>> 1 row(s) retrieved.
>>
>> Database closed.
>>
>> real 0m2.477s
>> user 0m0.012s
>> sys 0m0.002s
>>
>> Any thoughts on where I should start looking?
>>
>> Thanks,
>> -Justin
>>
>> ************************************************************
>> *******************
>>
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>> ************************************************************
>> *******************
>>
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>> ************************************************************
>> *******************
>>
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>> ************************************************************
>> *******************
>>
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>> ************************************************************
>> *******************
>> Forum Note: Use "Reply" to post a response in the discussion forum.
>>
>> --
>> Fernando Nunes
>> Portugal
>>
>> http://informix-technology.blogspot.com
>> My email works... but I don't check it frequently...
>>
>>
>>
>
*******************************************************************************
>> Forum Note: Use "Re
Yes, that is curious. I went poking around, and I found the root cause - a bad driver on the host OS: https://support.microsoft.com/en-us/help/2902166/poor-network-performance-on -virtual-machines-on-a-windows-server-2012 Iâve updated the driver on the host, and now everything runs as expected. Thanks everyone for your help. -Justin > On Feb 3, 2018, at 1:12 PM, Andreas Legner1 <Andreas.Legner1@de.ibm.com> wrote: > > Nor sure I'm reading this tcp trace right, but I'd say this this is an > exchange of about 12-14 packages in either way and it appears there's a > 150-200 ms delay for almost every reply from server to client while client > would react pretty much instantaneously. So no specific server operation > causing the majority of the overall 1.9 seconds duration obviously. > What I cannot tell is whether this delay occurs within the Informix server > or rather somewhere below it and within the network. > > Can we assume this trace has been collected on the client machine? What > would it look like when capturing on server side? > > What's the general load on the Informix server, what on its host? E.g. > wrt. cpu consumption? > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >