Re: Email performance/SMTP relays

[Dave Barber]

I've amended the mailbox polling intervals (reduce on one, increased on the
other), along with increasing the number of sender threads, outgoing
message queue size (to 500), but it has had no impact.

We're still finding that one mailbox is sending several hundred emails at
each polling interval (set to 4 minutes), and the other is only sending 3
or 4 emails every minute.

I am wondering if there are any restrictions on the mailbox itself, but I
have never been involved in exchange or mail relay operations.

Regards

Dave

On 27 April 2016 at 07:30, Dave Barber <daddy.bar...@gmail.com> wrote:

> Jason, thats a thought - I'll amend the polling interval on both of them -
> longer interval on the internal emails (timing isn't quite so important),
> shorter polling interval on the external mailbox.  Worth a try, then I'll
> look into the email properties as suggested by LJ.
>
> Ta.
>
> On 26 April 2016 at 16:41, Jason Miller <jason.mil...@gmail.com> wrote:
>
>> **
>> I also wonder if some timeout or number of message threshold is being hit
>> by the first mailbox and not leaving much processing room for the second
>> mailbox? By the time the first mailbox finishes sending there is little or
>> no threshold left for the second mailbox. If you know when a blast to
>> mailbox 2 is going to happen, as a simple test you could disable the first
>> mailbox (restart EE) and see if the blast in mailbox 2 processes quicker.
>>
>> Jason
>>
>> On Tue, Apr 26, 2016 at 8:33 AM, LJ LongWing <lj.longw...@gmail.com>
>> wrote:
>>
>>> **
>>> there are many settings you can set in the properties file to determine
>>> how many are sent per connection with the server, etcso, you may need
>>> to tweak some of those outgoing settings.
>>>
>>> On Tue, Apr 26, 2016 at 9:00 AM, Dave Barber <daddy.bar...@gmail.com>
>>> wrote:
>>>
>>>> **
>>>> I should have clarified - the discrepancy we're seeing is between the
>>>> create time and the time sent on the email messages form.  I have
>>>> monitoring in place that counts the number of unsent messages and sends me
>>>> an email (via the reliable gateway/mailbox) when there are more than a few
>>>> hundred unsent messages.
>>>>
>>>> Once they've been marked by the email messages form as "sent", they do
>>>> actually send fairly quick.  Possibly a problem at our end, but as we have
>>>> one mailbox that is sending without any performance problems, that is
>>>> making me think that the email engine itself is having no real problems.
>>>>
>>>> I cannot readily determine if the fault is somewhere on the network
>>>> between the AR Server and the gateway or the SMTP gateway itself.
>>>>
>>>> On 26 April 2016 at 13:31, Shellman, David <dave.shell...@te.com>
>>>> wrote:
>>>>
>>>>> **
>>>>>
>>>>> LJ,
>>>>>
>>>>>
>>>>>
>>>>> I agree.  Looking at the email headers will show where the delays
>>>>> occur.  Over the years I have seen delays within the email system.  I have
>>>>> also observed where the delay is handing off to the email server.  Reading
>>>>> the email heads was helpful in figuring it out.
>>>>>
>>>>>
>>>>>
>>>>> I also build a small two form system for monitoring email delays.  The
>>>>> first form was used to send an email that contained a GUID to our various
>>>>> outbound email addresses.  The email created record in the second form.
>>>>> Workflow on the second form used the GUID to update the Status of the
>>>>> record on the first form.  There were escalation notifications on the 
>>>>> first
>>>>> form that would send us an SMS if the record in the first form was not
>>>>> updated in 20 minutes.
>>>>>
>>>>>
>>>>>
>>>>> Dave
>>>>>
>>>>>
>>>>>
>>>>> *From:* Action Request System discussion list(ARSList) [mailto:
>>>>> arslist@ARSLIST.ORG] *On Behalf Of *LJ LongWing
>>>>> *Sent:* Tuesday, April 26, 2016 8:22 AM
>>>>> *To:* arslist@ARSLIST.ORG
>>>>> *Subject:* Re: Email performance/SMTP relays
>>>>>
>>>>>
>>>>>
>>>>> **
>>>>>
>>>>> Dave,
>>>>>
>>>>> The 

Re: Email performance/SMTP relays

[Dave Barber]

Jason, thats a thought - I'll amend the polling interval on both of them -
longer interval on the internal emails (timing isn't quite so important),
shorter polling interval on the external mailbox.  Worth a try, then I'll
look into the email properties as suggested by LJ.

Ta.

On 26 April 2016 at 16:41, Jason Miller <jason.mil...@gmail.com> wrote:

> **
> I also wonder if some timeout or number of message threshold is being hit
> by the first mailbox and not leaving much processing room for the second
> mailbox? By the time the first mailbox finishes sending there is little or
> no threshold left for the second mailbox. If you know when a blast to
> mailbox 2 is going to happen, as a simple test you could disable the first
> mailbox (restart EE) and see if the blast in mailbox 2 processes quicker.
>
> Jason
>
> On Tue, Apr 26, 2016 at 8:33 AM, LJ LongWing <lj.longw...@gmail.com>
> wrote:
>
>> **
>> there are many settings you can set in the properties file to determine
>> how many are sent per connection with the server, etcso, you may need
>> to tweak some of those outgoing settings.
>>
>> On Tue, Apr 26, 2016 at 9:00 AM, Dave Barber <daddy.bar...@gmail.com>
>> wrote:
>>
>>> **
>>> I should have clarified - the discrepancy we're seeing is between the
>>> create time and the time sent on the email messages form.  I have
>>> monitoring in place that counts the number of unsent messages and sends me
>>> an email (via the reliable gateway/mailbox) when there are more than a few
>>> hundred unsent messages.
>>>
>>> Once they've been marked by the email messages form as "sent", they do
>>> actually send fairly quick.  Possibly a problem at our end, but as we have
>>> one mailbox that is sending without any performance problems, that is
>>> making me think that the email engine itself is having no real problems.
>>>
>>> I cannot readily determine if the fault is somewhere on the network
>>> between the AR Server and the gateway or the SMTP gateway itself.
>>>
>>> On 26 April 2016 at 13:31, Shellman, David <dave.shell...@te.com> wrote:
>>>
>>>> **
>>>>
>>>> LJ,
>>>>
>>>>
>>>>
>>>> I agree.  Looking at the email headers will show where the delays
>>>> occur.  Over the years I have seen delays within the email system.  I have
>>>> also observed where the delay is handing off to the email server.  Reading
>>>> the email heads was helpful in figuring it out.
>>>>
>>>>
>>>>
>>>> I also build a small two form system for monitoring email delays.  The
>>>> first form was used to send an email that contained a GUID to our various
>>>> outbound email addresses.  The email created record in the second form.
>>>> Workflow on the second form used the GUID to update the Status of the
>>>> record on the first form.  There were escalation notifications on the first
>>>> form that would send us an SMS if the record in the first form was not
>>>> updated in 20 minutes.
>>>>
>>>>
>>>>
>>>> Dave
>>>>
>>>>
>>>>
>>>> *From:* Action Request System discussion list(ARSList) [mailto:
>>>> arslist@ARSLIST.ORG] *On Behalf Of *LJ LongWing
>>>> *Sent:* Tuesday, April 26, 2016 8:22 AM
>>>> *To:* arslist@ARSLIST.ORG
>>>> *Subject:* Re: Email performance/SMTP relays
>>>>
>>>>
>>>>
>>>> **
>>>>
>>>> Dave,
>>>>
>>>> The most important thing to figure out is where the delay isdo the
>>>> emails sit in your 'outbox' for up to 12 hours, or do they leave your mail
>>>> form in the normal timelines, and then simply take up to 12 hours to be
>>>> delivered?  I have found in this type of situation that the delay is
>>>> typically on the other end, not within Remedy.  Check that and if you find
>>>> that it reaches your smtp server in 'normal' time, then takes up to 12
>>>> hours to be delivered...then the admins of that smtp server need to
>>>> investigate where the delay is.
>>>>
>>>>
>>>>
>>>> On Tue, Apr 26, 2016 at 5:18 AM, Dave Barber <daddy.bar...@gmail.com>
>>>> wrote:
>>>>
>>>> **
>>>>
>>>> All,
>>>>
>>>> Experiencing performance problems on mail sending; some background ..

Re: Email performance/SMTP relays

[Jason Miller]

I also wonder if some timeout or number of message threshold is being hit
by the first mailbox and not leaving much processing room for the second
mailbox? By the time the first mailbox finishes sending there is little or
no threshold left for the second mailbox. If you know when a blast to
mailbox 2 is going to happen, as a simple test you could disable the first
mailbox (restart EE) and see if the blast in mailbox 2 processes quicker.

Jason

On Tue, Apr 26, 2016 at 8:33 AM, LJ LongWing <lj.longw...@gmail.com> wrote:

> **
> there are many settings you can set in the properties file to determine
> how many are sent per connection with the server, etcso, you may need
> to tweak some of those outgoing settings.
>
> On Tue, Apr 26, 2016 at 9:00 AM, Dave Barber <daddy.bar...@gmail.com>
> wrote:
>
>> **
>> I should have clarified - the discrepancy we're seeing is between the
>> create time and the time sent on the email messages form.  I have
>> monitoring in place that counts the number of unsent messages and sends me
>> an email (via the reliable gateway/mailbox) when there are more than a few
>> hundred unsent messages.
>>
>> Once they've been marked by the email messages form as "sent", they do
>> actually send fairly quick.  Possibly a problem at our end, but as we have
>> one mailbox that is sending without any performance problems, that is
>> making me think that the email engine itself is having no real problems.
>>
>> I cannot readily determine if the fault is somewhere on the network
>> between the AR Server and the gateway or the SMTP gateway itself.
>>
>> On 26 April 2016 at 13:31, Shellman, David <dave.shell...@te.com> wrote:
>>
>>> **
>>>
>>> LJ,
>>>
>>>
>>>
>>> I agree.  Looking at the email headers will show where the delays
>>> occur.  Over the years I have seen delays within the email system.  I have
>>> also observed where the delay is handing off to the email server.  Reading
>>> the email heads was helpful in figuring it out.
>>>
>>>
>>>
>>> I also build a small two form system for monitoring email delays.  The
>>> first form was used to send an email that contained a GUID to our various
>>> outbound email addresses.  The email created record in the second form.
>>> Workflow on the second form used the GUID to update the Status of the
>>> record on the first form.  There were escalation notifications on the first
>>> form that would send us an SMS if the record in the first form was not
>>> updated in 20 minutes.
>>>
>>>
>>>
>>> Dave
>>>
>>>
>>>
>>> *From:* Action Request System discussion list(ARSList) [mailto:
>>> arslist@ARSLIST.ORG] *On Behalf Of *LJ LongWing
>>> *Sent:* Tuesday, April 26, 2016 8:22 AM
>>> *To:* arslist@ARSLIST.ORG
>>> *Subject:* Re: Email performance/SMTP relays
>>>
>>>
>>>
>>> **
>>>
>>> Dave,
>>>
>>> The most important thing to figure out is where the delay isdo the
>>> emails sit in your 'outbox' for up to 12 hours, or do they leave your mail
>>> form in the normal timelines, and then simply take up to 12 hours to be
>>> delivered?  I have found in this type of situation that the delay is
>>> typically on the other end, not within Remedy.  Check that and if you find
>>> that it reaches your smtp server in 'normal' time, then takes up to 12
>>> hours to be delivered...then the admins of that smtp server need to
>>> investigate where the delay is.
>>>
>>>
>>>
>>> On Tue, Apr 26, 2016 at 5:18 AM, Dave Barber <daddy.bar...@gmail.com>
>>> wrote:
>>>
>>> **
>>>
>>> All,
>>>
>>> Experiencing performance problems on mail sending; some background ...
>>>
>>> Server is running ITSM 7.6 Change management on a 7.5 Solaris server,
>>> Oracle at the back end.  General application performance is fine.
>>>
>>> The server is making use of 2 outgoing mailboxes.
>>>
>>> Mailbox 1 - used for system notifications, such as "This change requires
>>> your approval", etc.  connects to a "legacy" SMTP relay, up to a few
>>> thousand emails sent each day, runs fine.  Emails are picked up/sent within
>>> a minute or two.
>>>
>>> Mailbox 2 - used for a custom notifications system (to notify customers
>>> of outages), similarly sends a couple of thousand emails per day.  Has

Re: Email performance/SMTP relays

[LJ LongWing]

there are many settings you can set in the properties file to determine how
many are sent per connection with the server, etcso, you may need to
tweak some of those outgoing settings.

On Tue, Apr 26, 2016 at 9:00 AM, Dave Barber <daddy.bar...@gmail.com> wrote:

> **
> I should have clarified - the discrepancy we're seeing is between the
> create time and the time sent on the email messages form.  I have
> monitoring in place that counts the number of unsent messages and sends me
> an email (via the reliable gateway/mailbox) when there are more than a few
> hundred unsent messages.
>
> Once they've been marked by the email messages form as "sent", they do
> actually send fairly quick.  Possibly a problem at our end, but as we have
> one mailbox that is sending without any performance problems, that is
> making me think that the email engine itself is having no real problems.
>
> I cannot readily determine if the fault is somewhere on the network
> between the AR Server and the gateway or the SMTP gateway itself.
>
> On 26 April 2016 at 13:31, Shellman, David <dave.shell...@te.com> wrote:
>
>> **
>>
>> LJ,
>>
>>
>>
>> I agree.  Looking at the email headers will show where the delays occur.
>> Over the years I have seen delays within the email system.  I have also
>> observed where the delay is handing off to the email server.  Reading the
>> email heads was helpful in figuring it out.
>>
>>
>>
>> I also build a small two form system for monitoring email delays.  The
>> first form was used to send an email that contained a GUID to our various
>> outbound email addresses.  The email created record in the second form.
>> Workflow on the second form used the GUID to update the Status of the
>> record on the first form.  There were escalation notifications on the first
>> form that would send us an SMS if the record in the first form was not
>> updated in 20 minutes.
>>
>>
>>
>> Dave
>>
>>
>>
>> *From:* Action Request System discussion list(ARSList) [mailto:
>> arslist@ARSLIST.ORG] *On Behalf Of *LJ LongWing
>> *Sent:* Tuesday, April 26, 2016 8:22 AM
>> *To:* arslist@ARSLIST.ORG
>> *Subject:* Re: Email performance/SMTP relays
>>
>>
>>
>> **
>>
>> Dave,
>>
>> The most important thing to figure out is where the delay isdo the
>> emails sit in your 'outbox' for up to 12 hours, or do they leave your mail
>> form in the normal timelines, and then simply take up to 12 hours to be
>> delivered?  I have found in this type of situation that the delay is
>> typically on the other end, not within Remedy.  Check that and if you find
>> that it reaches your smtp server in 'normal' time, then takes up to 12
>> hours to be delivered...then the admins of that smtp server need to
>> investigate where the delay is.
>>
>>
>>
>> On Tue, Apr 26, 2016 at 5:18 AM, Dave Barber <daddy.bar...@gmail.com>
>> wrote:
>>
>> **
>>
>> All,
>>
>> Experiencing performance problems on mail sending; some background ...
>>
>> Server is running ITSM 7.6 Change management on a 7.5 Solaris server,
>> Oracle at the back end.  General application performance is fine.
>>
>> The server is making use of 2 outgoing mailboxes.
>>
>> Mailbox 1 - used for system notifications, such as "This change requires
>> your approval", etc.  connects to a "legacy" SMTP relay, up to a few
>> thousand emails sent each day, runs fine.  Emails are picked up/sent within
>> a minute or two.
>>
>> Mailbox 2 - used for a custom notifications system (to notify customers
>> of outages), similarly sends a couple of thousand emails per day.  Has been
>> in active use since February this year.  As it is external emails it is
>> using a "mass mail" SMTP relay designated for this purpose.
>>
>> It is this second one that is having problems - emails take anything from
>> 3 minutes to 12 (!!) hours to be sent.
>>
>> As far as I can see the basic configuration/polling is setup the same on
>> both (2 minutes).
>>
>> Any suggestions?  Would reducing the polling interval have a positive
>> change?  Or are there further configuration changes on the email engine
>> itself that could improve performance?  Or am I just at the mercy of a
>> sluggish SMTP relay?
>>
>> (I've never had to investigate email performance problems!)
>>
>> Regards
>>
>> Dave
>>
>> _ARSlist: "Where the Answers Are" and have been for 20 years_
>>
>>
>>
>> _ARSlist: "Where the Answers Are" and have been for 20 years_
>> _ARSlist: "Where the Answers Are" and have been for 20 years_
>>
>
> _ARSlist: "Where the Answers Are" and have been for 20 years_
>

___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
"Where the Answers Are, and have been for 20 years"

Re: Email performance/SMTP relays

[Dave Barber]

I should have clarified - the discrepancy we're seeing is between the
create time and the time sent on the email messages form.  I have
monitoring in place that counts the number of unsent messages and sends me
an email (via the reliable gateway/mailbox) when there are more than a few
hundred unsent messages.

Once they've been marked by the email messages form as "sent", they do
actually send fairly quick.  Possibly a problem at our end, but as we have
one mailbox that is sending without any performance problems, that is
making me think that the email engine itself is having no real problems.

I cannot readily determine if the fault is somewhere on the network between
the AR Server and the gateway or the SMTP gateway itself.

On 26 April 2016 at 13:31, Shellman, David <dave.shell...@te.com> wrote:

> **
>
> LJ,
>
>
>
> I agree.  Looking at the email headers will show where the delays occur.
> Over the years I have seen delays within the email system.  I have also
> observed where the delay is handing off to the email server.  Reading the
> email heads was helpful in figuring it out.
>
>
>
> I also build a small two form system for monitoring email delays.  The
> first form was used to send an email that contained a GUID to our various
> outbound email addresses.  The email created record in the second form.
> Workflow on the second form used the GUID to update the Status of the
> record on the first form.  There were escalation notifications on the first
> form that would send us an SMS if the record in the first form was not
> updated in 20 minutes.
>
>
>
> Dave
>
>
>
> *From:* Action Request System discussion list(ARSList) [mailto:
> arslist@ARSLIST.ORG] *On Behalf Of *LJ LongWing
> *Sent:* Tuesday, April 26, 2016 8:22 AM
> *To:* arslist@ARSLIST.ORG
> *Subject:* Re: Email performance/SMTP relays
>
>
>
> **
>
> Dave,
>
> The most important thing to figure out is where the delay isdo the
> emails sit in your 'outbox' for up to 12 hours, or do they leave your mail
> form in the normal timelines, and then simply take up to 12 hours to be
> delivered?  I have found in this type of situation that the delay is
> typically on the other end, not within Remedy.  Check that and if you find
> that it reaches your smtp server in 'normal' time, then takes up to 12
> hours to be delivered...then the admins of that smtp server need to
> investigate where the delay is.
>
>
>
> On Tue, Apr 26, 2016 at 5:18 AM, Dave Barber <daddy.bar...@gmail.com>
> wrote:
>
> **
>
> All,
>
> Experiencing performance problems on mail sending; some background ...
>
> Server is running ITSM 7.6 Change management on a 7.5 Solaris server,
> Oracle at the back end.  General application performance is fine.
>
> The server is making use of 2 outgoing mailboxes.
>
> Mailbox 1 - used for system notifications, such as "This change requires
> your approval", etc.  connects to a "legacy" SMTP relay, up to a few
> thousand emails sent each day, runs fine.  Emails are picked up/sent within
> a minute or two.
>
> Mailbox 2 - used for a custom notifications system (to notify customers of
> outages), similarly sends a couple of thousand emails per day.  Has been in
> active use since February this year.  As it is external emails it is using
> a "mass mail" SMTP relay designated for this purpose.
>
> It is this second one that is having problems - emails take anything from
> 3 minutes to 12 (!!) hours to be sent.
>
> As far as I can see the basic configuration/polling is setup the same on
> both (2 minutes).
>
> Any suggestions?  Would reducing the polling interval have a positive
> change?  Or are there further configuration changes on the email engine
> itself that could improve performance?  Or am I just at the mercy of a
> sluggish SMTP relay?
>
> (I've never had to investigate email performance problems!)
>
> Regards
>
> Dave
>
> _ARSlist: "Where the Answers Are" and have been for 20 years_
>
>
>
> _ARSlist: "Where the Answers Are" and have been for 20 years_
> _ARSlist: "Where the Answers Are" and have been for 20 years_
>

___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
"Where the Answers Are, and have been for 20 years"

Re: Email performance/SMTP relays

[Shellman, David]

LJ,

I agree.  Looking at the email headers will show where the delays occur.  Over 
the years I have seen delays within the email system.  I have also observed 
where the delay is handing off to the email server.  Reading the email heads 
was helpful in figuring it out.

I also build a small two form system for monitoring email delays.  The first 
form was used to send an email that contained a GUID to our various outbound 
email addresses.  The email created record in the second form.  Workflow on the 
second form used the GUID to update the Status of the record on the first form. 
 There were escalation notifications on the first form that would send us an 
SMS if the record in the first form was not updated in 20 minutes.

Dave

From: Action Request System discussion list(ARSList) 
[mailto:arslist@ARSLIST.ORG] On Behalf Of LJ LongWing
Sent: Tuesday, April 26, 2016 8:22 AM
To: arslist@ARSLIST.ORG
Subject: Re: Email performance/SMTP relays

**
Dave,
The most important thing to figure out is where the delay isdo the emails 
sit in your 'outbox' for up to 12 hours, or do they leave your mail form in the 
normal timelines, and then simply take up to 12 hours to be delivered?  I have 
found in this type of situation that the delay is typically on the other end, 
not within Remedy.  Check that and if you find that it reaches your smtp server 
in 'normal' time, then takes up to 12 hours to be delivered...then the admins 
of that smtp server need to investigate where the delay is.

On Tue, Apr 26, 2016 at 5:18 AM, Dave Barber 
<daddy.bar...@gmail.com<mailto:daddy.bar...@gmail.com>> wrote:
**
All,
Experiencing performance problems on mail sending; some background ...
Server is running ITSM 7.6 Change management on a 7.5 Solaris server, Oracle at 
the back end.  General application performance is fine.
The server is making use of 2 outgoing mailboxes.
Mailbox 1 - used for system notifications, such as "This change requires your 
approval", etc.  connects to a "legacy" SMTP relay, up to a few thousand emails 
sent each day, runs fine.  Emails are picked up/sent within a minute or two.
Mailbox 2 - used for a custom notifications system (to notify customers of 
outages), similarly sends a couple of thousand emails per day.  Has been in 
active use since February this year.  As it is external emails it is using a 
"mass mail" SMTP relay designated for this purpose.
It is this second one that is having problems - emails take anything from 3 
minutes to 12 (!!) hours to be sent.
As far as I can see the basic configuration/polling is setup the same on both 
(2 minutes).
Any suggestions?  Would reducing the polling interval have a positive change?  
Or are there further configuration changes on the email engine itself that 
could improve performance?  Or am I just at the mercy of a sluggish SMTP relay?
(I've never had to investigate email performance problems!)
Regards
Dave
_ARSlist: "Where the Answers Are" and have been for 20 years_

_ARSlist: "Where the Answers Are" and have been for 20 years_

___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
"Where the Answers Are, and have been for 20 years"

Re: Email performance/SMTP relays

[LJ LongWing]

Dave,
The most important thing to figure out is where the delay isdo the
emails sit in your 'outbox' for up to 12 hours, or do they leave your mail
form in the normal timelines, and then simply take up to 12 hours to be
delivered?  I have found in this type of situation that the delay is
typically on the other end, not within Remedy.  Check that and if you find
that it reaches your smtp server in 'normal' time, then takes up to 12
hours to be delivered...then the admins of that smtp server need to
investigate where the delay is.

On Tue, Apr 26, 2016 at 5:18 AM, Dave Barber  wrote:

> **
> All,
>
> Experiencing performance problems on mail sending; some background ...
>
> Server is running ITSM 7.6 Change management on a 7.5 Solaris server,
> Oracle at the back end.  General application performance is fine.
>
> The server is making use of 2 outgoing mailboxes.
>
> Mailbox 1 - used for system notifications, such as "This change requires
> your approval", etc.  connects to a "legacy" SMTP relay, up to a few
> thousand emails sent each day, runs fine.  Emails are picked up/sent within
> a minute or two.
>
> Mailbox 2 - used for a custom notifications system (to notify customers of
> outages), similarly sends a couple of thousand emails per day.  Has been in
> active use since February this year.  As it is external emails it is using
> a "mass mail" SMTP relay designated for this purpose.
>
> It is this second one that is having problems - emails take anything from
> 3 minutes to 12 (!!) hours to be sent.
>
> As far as I can see the basic configuration/polling is setup the same on
> both (2 minutes).
>
> Any suggestions?  Would reducing the polling interval have a positive
> change?  Or are there further configuration changes on the email engine
> itself that could improve performance?  Or am I just at the mercy of a
> sluggish SMTP relay?
>
> (I've never had to investigate email performance problems!)
>
> Regards
>
> Dave
> _ARSlist: "Where the Answers Are" and have been for 20 years_

___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
"Where the Answers Are, and have been for 20 years"

Email performance/SMTP relays

[Dave Barber]

All,

Experiencing performance problems on mail sending; some background ...

Server is running ITSM 7.6 Change management on a 7.5 Solaris server,
Oracle at the back end.  General application performance is fine.

The server is making use of 2 outgoing mailboxes.

Mailbox 1 - used for system notifications, such as "This change requires
your approval", etc.  connects to a "legacy" SMTP relay, up to a few
thousand emails sent each day, runs fine.  Emails are picked up/sent within
a minute or two.

Mailbox 2 - used for a custom notifications system (to notify customers of
outages), similarly sends a couple of thousand emails per day.  Has been in
active use since February this year.  As it is external emails it is using
a "mass mail" SMTP relay designated for this purpose.

It is this second one that is having problems - emails take anything from 3
minutes to 12 (!!) hours to be sent.

As far as I can see the basic configuration/polling is setup the same on
both (2 minutes).

Any suggestions?  Would reducing the polling interval have a positive
change?  Or are there further configuration changes on the email engine
itself that could improve performance?  Or am I just at the mercy of a
sluggish SMTP relay?

(I've never had to investigate email performance problems!)

Regards

Dave

___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
"Where the Answers Are, and have been for 20 years"