mode=3 assistance

[rossi2000]

afternoon folks
i cant seem to grasp the tutorial up at the top of this forum by lenny on mode3 cachex.

can anyone help with the below example?, its mainly the group assignment bits.

(ME)
readers:-
reader a = local = group 1
reader b = share = group 2
reader c = share = group 3
reader d = cachex3 = group ?

users:-
user a = me = group 1,2,3
user b = me2 = group 1,2,3
user c = cachex3 = group ?

(FREIND)
readers:-
reader a = local = group 1
reader b = share = group 2
reader c = cachex3 = group ?

users:-
user a = me = group 1,2
user b = me2 = group 1,2
user c = cachex3 = group ?


we both want to exchange cache on the local cards only.
should it be like this?

(ME)
readers:-
reader a = local = group 1
reader b = share = group 2
reader c = share = group 3
reader d = cachex3 = group 1

users:-
user a = me = group 1,2,3,20
user b = me2 = group 1,2,3,20
user c = cachex3 = group 20

(FREIND)
readers:-
reader a = local = group 1
reader b = share = group 2
reader c = cachex3 = group 1

users:-
user a = me = group 1,2,20
user b = me2 = group 1,2,20
user c = cachex3 = group 20


cheers

[turbopower]

Hello read this
http://www.satsupreme.com/showthread...l=1#post929864

[Qnapper]

I have cache 3 running no problem with 6 peers. Here is how it works:

The [reader] becomes the user and the [User] becomes the reader...... so to set it up right you need to do this:

[account]
user = user1
pwd = user1
description = Cache 3 user 2
disabled = 0
group = 7

[account]
user = user2
pwd = user2
description = Cache 3 user 2
disabled = 0
group = 8

[account]
user = user3
pwd = user3
description = Cache 3 user 3
disabled = 0
group = 9

[account]
user = user4
pwd = user4
description = Cache 3 user 4
disabled = 0
group = 10

---------------------------------------

[reader]
label = user1
description =
enable = 1
group = 8,9,10

[reader]
label = user2
description =
enable = 1
group = 7,9,10

[reader]
label = user3
description =
enable = 1
group = 7,8,10

[reader]
label = user4
description =
enable = 1
group =7,8,9

Remember the [reader] is now SENDING the cache so you NEVER send him his OWN group....

Now:

I have found that if I add the above groups to my NORMAL peers ie: 7,8,9,10 that they get ecm from my cache first every time (if I have it) and shows as (cache3) before reading my local card.

I hope this helps you, I cannot take credit for this as a friend of mine ( Mark you know who you are ) spent 3 days tuting me to get this right.

Napper

[CapNCooK]

Well.., for the most part you are right

Altough, I don't understand why you use al these groups, other than you want to prevent 'loopback' from the same user to the same reader..

But in mode-3 and mode-1, loopback does not exist, so you can put all cache- readers/users into the same groups.
All your normal (non-cache) users, should have (at least) the same readergroup assigned of all cache-ex users.

--

Therefore, i advise to let the groups in all cache-users/readers overlap the groups you are using with the normal readers.

For example:


[account]
user = user1
pwd = user1
description = SKY UK CARD USER WITH CACHE-EX SUPPORT
disabled = 0
group = 1

[account]
user = user2
pwd = user2
description = SKY IT CARD USER WITH CACHE-EX SUPPORT
disabled = 0
group = 2

[account]
user = user3
pwd = user3
description = SKY IT / UK CARD USER WITH CACHE-EX SUPPORT
disabled = 0
group = 1,2

[account]
user = user4
pwd = user4
description = CACHE-EX USER 1 - Mode-3
disabled = 0
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

[account]
user = user5
pwd = user5
description = CACHE-EX USER 2 - Mode-3
disabled = 0
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

---------------------------------------

[reader]
label = reader1
description = REAL CARD READER SKY UK
enable = 1
group = 1

[reader]
label = reader2
description = REAL CARD READER SKY IT
enable = 1
group = 2

[reader]
label = reader3
description = CACHE-EX READER 1 - Mode3
enable = 1
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

[reader]
label = user4
description = CACHE-EX READER 2 - Mode3
enable = 1
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

Hope this helps!

[Qnapper]

Well.., for the most part you are right

Altough, I don't understand why you use al these groups, other than you want to prevent 'loopback' from the same user to the same reader..

But in mode-3 and mode-1, loopback does not exist, so you can put all cache- readers/users into the same groups.
All your normal (non-cache) users, should have (at least) the same readergroup assigned of all cache-ex users.

--

Therefore, i advise to let the groups in all cache-users/readers overlap the groups you are using with the normal readers.

For example:


[account]
user = user1
pwd = user1
description = SKY UK CARD USER WITH CACHE-EX SUPPORT
disabled = 0
group = 1

[account]
user = user2
pwd = user2
description = SKY IT CARD USER WITH CACHE-EX SUPPORT
disabled = 0
group = 2

[account]
user = user3
pwd = user3
description = SKY IT / UK CARD USER WITH CACHE-EX SUPPORT
disabled = 0
group = 1,2

[account]
user = user4
pwd = user4
description = CACHE-EX USER 1 - Mode-3
disabled = 0
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

[account]
user = user5
pwd = user5
description = CACHE-EX USER 2 - Mode-3
disabled = 0
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

---------------------------------------

[reader]
label = reader1
description = REAL CARD READER SKY UK
enable = 1
group = 1

[reader]
label = reader2
description = REAL CARD READER SKY IT
enable = 1
group = 2

[reader]
label = reader3
description = CACHE-EX READER 1 - Mode3
enable = 1
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

[reader]
label = user4
description = CACHE-EX READER 2 - Mode3
enable = 1
cacheex=3
group = 1,2 # <= ALL GROUPS YOU USE IN AL NORMAL/PROXY READERS

Hope this helps!

Most confused now as for 3 days we ( Mark and I ) tried all ways and got loop backs - that's why I showed my post in colour as it works 100% perfect and has been running now for nearly 4 days without problems. I see that you have added Sky IT and Sky UK - I take it that is for filtering which peer has local card in cache 3? - would have thought that exchanging cache does not need such filters.

Thank you by the way for coming up and helping here. There are too many people doing this wrong and causing problems for others. Without your knowledge we would be lost in a sea of cache.....

Napper

[CapNCooK]

Most confused now as for 3 days we ( Mark and I ) tried all ways and got loop backs

Well, don't get too confused, you understand the concept of mode-3 perfectly i think.
Its just the groups assignment you made because of the detected 'loopbacks' - which do not exist - that's concerning me. :)

--

How do you define and/or detected these 'loopbacks' ?
Do you mean your RAM is filling up ?

About the SKY IT/UK. I tried to represent a mix of two 'normal' readers with two real / different cards, and two cache readers/users in my example above, and they has nothing to do with filtering or such.

So in the example, i put in:
user 1 - REAL user (dreambox or such), with accessrights to use reader1, making use of your cache-ex cache (because cache is also in group 1,2).
user 2 - REAL user (dreambox or such), with accessrights to use reader2, making use of your cache-ex cache (because cache is also in group 1,2).
user 3 - REAL user (dreambox or such), with accessrights to use reader1 and reader2, making use of your cache-ex cache (because cache is made available for group 1,2)
user 4 - CACHE-EX user 1, receiving cache, and making that cache accessable for all peers in group 1,2
user 5 - CACHE-EX user 2, receiving cache, and making that cache accessable for all peers in group 1,2

reader 1 - REAL cardreader 1 , with real card inserted
reader 2 - REAL cardreader 2 , with real card inserted
reader 3 - CACHE-EX reader 1, transmitting all cache received by and/or generated in group 1,2
reader 4 - CACHE-EX reader 2, transmitting all cache received by and/or generated in group 1,2

Hope this helps!

[Qnapper]

It is indeed ram filling up. I have now set all cache peers to hop4 , which appears to be perfect. I have 2GB ram and using ihop can now see MEM% is around the 1.5 mark 247, were as before it was up to 12% at times. I thought it was memory leak, but turned out to be filling of the ram from cache.

Napper

[CapNCooK]

It is indeed ram filling up. I have now set all cache peers to hop4 , which appears to be perfect. I have 2GB ram and using ihop can now see MEM% is around the 1.5 mark 247, were as before it was up to 12% at times. I thought it was memory leak, but turned out to be filling of the ram from cache.

Napper

This is a memory leak / bug in oscam, and has nothing to do with loopbacks.

See here: http://www.satsupreme.com/showthread...with-cacheex-3
And here: http://streamboard.de.vu:8001/ticket/2676

Revert to 7190, as it is the latest stable build for me with cache-ex
(although 7190 still uses a little more CPU, but that's no problem for me)

ThX

[Qnapper]

Will have a look at build 7190, currently using 7180 for past 4/5 days seams fine.

Thanks again I´m sure many will read the above with interest.

Napper

[turbopower]

If you want to reduce cpu usage and provide stable connection on your peers always set in global config.

clientmaxidle = 0