best cach config require

[[-SONIC-]]

Hello, I hope you can help me.
I'm sorry for my bad english: (

I have Oscam it on a VPS server.
It run CacheEX, Proxy & Client on the server. At home I have a Home Server because only my Local maps are on it cs378x go over to the VPS server.

I use mainly CacheEX Mode 2 (CS378x) and rarely Mode 3 (CCcam)

Unfortunately I have the problem, for example on Sky Italia (CAID 09CD Although I use this very often have the Share and cache and also local) I look such as a channel, with super ECM times. Suddenly, the picture will freeze easily and after 5 seconds it goes on again. In the log there is nothing to see! In what may lie? Use the last Oscam versions! Wichtig für mich sind CAID 09CD, 1702(1833), 1830, 183D

[global]
logduplicatelines             = 1
usrfileflag                   = 1
clienttimeout                 = 5500
fallbacktimeout               = 5500
netprio                       = 1
unlockparental                = 1
nice                          = -4
maxlogsize                    = 512
waitforcards                  = 0
waitforcards_extra_delay      = 0
preferlocalcards              = 1
dropdups                      = 1
block_same_ip                 = 0
block_same_name               = 0
emmlogdir                     = /var/log/
lb_mode                       = 1
lb_save                       = 1600
lb_min_ecmcount               = 10
lb_max_ecmcount               = 1500
lb_retrylimit                 = 2800
lb_max_readers                = 2
lb_auto_betatunnel            = 0
lb_auto_betatunnel_prefer_beta= 0
lb_retrylimits                = 0963:440,09CD:430,093B:430,0919:430,098C:430,09C4:430,09C7:600,090F:2050,093E:1850,0100:4200,0500:3850,0B00:3200,0B01:3200,0B02:1000,0BAA:1400,17:5500
lb_nbest_percaid              = 09:2
lb_noproviderforcaid          = 0D,17,18
ecmfmt                        = c&p/i/s/l:h
failbantime                   = 1440

[cache]
delay                         = 10
max_hit_time                  = 3
wait_time                     = 0100:250,0500:500,06:280,09:25,0B02:250,0B:500,0D:200,1702:950,1722:1350,1830:600,1843:750,183D:120,18:550
csp_allow_reforward           = 1
cwcycle_check_enable          = 1
cwcycle_check_caid            = 0100,0500,0648,09C7,0B00,0B01,0B02,0D05,0D95,0D96,1702,1722,1803,1810,1830,1833,1834,1835,1838,183D,1843
cwcycle_maxlist               = 4000
cwcycle_allowbadfromffb       = 1
cwcycle_usecwcfromce          = 1

THX 4 THE HELP!

[CapNCooK]

Enable CWC for 09CD in latest SVN, and lower NDS waittimes..

[[-SONIC-]]

Hello and thank you for your answer.

Do you mean that I have to do it like this:

cwcycle_check_caid = 09CD
wait_time = 09CD: 200


Preferlocalcards = 0,1 or 2 ???????

Thanks for your help

[PlatoonRTV]

My hits also lowered with new oscam. I use cacheex oscam on different location and altered settings dont help. I increased incoming hops to 5, changed time values to 60 but hits are still low on my second oscam at home (old version). When i use old version on my cacheex oscam, hits return to normal. I really dont understand this, should be better with new oscam right?

[DFC90]

yesterday I have my server completely overhauled and provided with a completely new concept. now I have 91% cache hits and produce a lot of local cache!

[acac1979]

yesterday I have my server completely overhauled and provided with a completely new concept. now I have 91% cache hits and produce a lot of local cache!

Hi,
I think it is important, each user share your experience with cacheex putting your current settings, so we can find out what settings that optimize and maximize the hits!

At this moment iam testing the cacheex mod3 with oscam users, and i have this settings. Only caids 1802 and 0100 are of my interest. Any opinion are welcome!!

[cache]
max_time = 60
max_hit_time = 60
wait_time = 0:10000:0,1802:1500:0,0100:1500:0
csp_allow_request = 0
cacheex_cw_check = 0:0:2, 1802:0:1, 0100:0:1
cwcycle_check_enable = 1
cwcycle_check_caid = 1802,0100
cwcycle_maxlist = 4000
cwcycle_keeptime = 240
cwcycle_usecwcfromce = 1


[reader]
label =
protocol = cccam
device =
user = xxxx
password = yyyy
reconnecttimeout = 60
cacheex = 3
cacheex_maxhop = 2
group = 2,3,4,6,7
ccckeepalive = 1
fallback = 1

[account]
user = xxxx
pwd = yyyy
group = 4
cacheex = 3
cacheex_maxhop = 5
cccmaxhops = -1
uniq = 3
keepalive = 1
cacheex_drop_csp = 1
caid = 1802,0100
cccreshare = 0

With this settings 75% of my requests are answered by cache!

I don´t know if it is recommended lower the value of the wait time (1500 ms). What is your opinion? If i set the wait time to 600ms, for example, i obtain a very high number of wai_time _overs, and i don´t know if when this information appear, i obtained a "hit" really?
Further, when i don´t have the ECM request in cacheex, they will be opened through my local server or proxys, with values ​​of 1700-1800ms, which can be a disadvantage!

I am using oscam svn r9142

Best regards

[galiano]

@acac1979
Lately I see a tendency to set the huge waiting times, which vitiates the idea of cacheex.
Just a rhetorical question:
If any of your ordinary peers (no cache), see that your card response times are 1700-1800ms., what would he think?

I think the idea of cacheex is, to be an aid to local card and not the main way of watching TV!

[CapNCooK]

I think it is important, each user share your experience with cacheex putting your current settings, so we can find out what settings that optimize and maximize the hits! At this moment iam testing the cacheex mod3 with oscam users, and i have this settings. Only caids 1802 and 0100 are of my interest. Any opinion are welcome!!

Please update to latest SVN as a bug is resolved yesterday. Also CWC has some flaws currently, will be updated soon in SVN.
For now/test, disable CWC and try to set waittime like: wait_time = 0:300:450,1802:800:1500,0100:800:1200

If i set the wait time to 600ms, for example, i obtain a very high number of wai_time _overs, and i don´t know if when this information appear, i obtained a "hit" really?

Well if it "hits" depends on your peers, but oscam has sent the request to normal cards/proxy.. yes.

Further, when i don´t have the ECM request in cacheex, they will be opened through my local server or proxys, with values ​​of 1700-1800ms, which can be a disadvantage!

True, but this is where dynamic wait comes into play. See my waittime specification above.. Example: 1802:800:1500, means it will wait for 800ms by default, but if a cache3 hit follows.. it switches to dynwait of 1500ms. (until a normal hit follows, then it switches back).

[CapNCooK]

If any of your ordinary peers (no cache), see that your card response times are 1700-1800ms., what would he think?

He should think you are using cache.. and therefore a CW could be late(r).
Not really a problem if the CW is "on-time" for the given CAID ?

[acac1979]

Please update to latest SVN as a bug is resolved yesterday. Also CWC has some flaws currently, will be updated soon in SVN.
For now/test, disable CWC and try to set waittime like: wait_time = 0:300:450,1802:800:1500,0100:800:1200


Well if it "hits" depends on your peers, but oscam has sent the request to normal cards/proxy.. yes.


True, but this is where dynamic wait comes into play. See my waittime specification above.. Example: 1802:800:1500, means it will wait for 800ms by default, but if a cache3 hit follows.. it switches to dynwait of 1500ms. (until a normal hit follows, then it switches back).

Thanks CapNCooK,

Now i updated my svn for 9144. I disabled also the cycle check and added the wait time 0:300:450,1802:800:1500,0100:800:1200

Lets see in next 24h the behavior of cacheex!

Best regards