kernel varoittelee

[Pakana]

Koodia: [Valitse]

Feb 15 23:42:37 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:42:38 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:42:39 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:42:40 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:42:41 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:42:51 pakanana syslog-ng[2289]: last message repeated 9 times
Feb 15 23:42:51 pakanana kernel: EDAC MC0: CE page 0x0, offset 0x0, grain 1073741824, syndrome 0x0, row 2, channel 0, label "": AMD762
Feb 15 23:42:52 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:42:53 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:42:54 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:42:54 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:42:56 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:43:25 pakanana syslog-ng[2289]: last message repeated 28 times
Feb 15 23:43:25 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:43:26 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:43:38 pakanana syslog-ng[2289]: last message repeated 11 times
Feb 15 23:43:38 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:43:39 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 15 23:43:43 pakanana syslog-ng[2289]: last message repeated 3 times
Feb 15 23:43:43 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:43:43 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 15 23:43:45 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762

Tuossa pieni näyte. Mitä piruja nuo ovat?

[jmp]

Onkohan muistit kunnossa?

Kokeile ajaa memtest86 jokun aikaa.

[Pakana]

Tässä taas pätkä, tällä kertaa /var/log/messages fileestä.
Tätä siis tulee kokoajan, luulisi että järjestelmä hiettää kupit jos muisteissa noin paljon virheitä olisi... Tosin muisti ovat kyllä ecc palikoita.

Koodia: [Valitse]

Feb 16 22:28:24 pakanana kernel: EDAC MC0: CE page 0x0, offset 0x0, grain 1073741824, syndrome 0x0, row 2, channel 0, label "": AMD762
Feb 16 22:28:25 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:25 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:27 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:27 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 0.
Feb 16 22:28:27 pakanana kernel: Bank 1: d400400000000853
Feb 16 22:28:27 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 0.
Feb 16 22:28:27 pakanana kernel: Bank 2: d400400000000813
Feb 16 22:28:27 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 1.
Feb 16 22:28:27 pakanana kernel: Bank 1: d400400000000853
Feb 16 22:28:27 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 1.
Feb 16 22:28:27 pakanana kernel: Bank 2: d400400000000813
Feb 16 22:28:28 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:34 pakanana syslog-ng[2289]: last message repeated 5 times
Feb 16 22:28:34 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 16 22:28:35 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:36 pakanana kernel: EDAC MC0: CE page 0x0, offset 0x0, grain 1073741824, syndrome 0x0, row 2, channel 0, label "": AMD762
Feb 16 22:28:37 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:37 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:39 pakanana kernel: EDAC MC0: CE page 0x60000, offset 0x0, grain 536870912, syndrome 0x0, row 1, channel 0, label "": AMD762
Feb 16 22:28:40 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:42 pakanana syslog-ng[2289]: last message repeated 2 times
Feb 16 22:28:42 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 0.
Feb 16 22:28:42 pakanana kernel: Bank 1: d400400000000853
Feb 16 22:28:42 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 0.
Feb 16 22:28:42 pakanana kernel: Bank 2: d400400000000813
Feb 16 22:28:42 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 1.
Feb 16 22:28:42 pakanana kernel: Bank 1: d400400000000853
Feb 16 22:28:42 pakanana kernel: MCE: The hardware reports a non fatal, correctable incident occurred on CPU 1.
Feb 16 22:28:42 pakanana kernel: Bank 2: d400400000000813
Feb 16 22:28:43 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:46 pakanana syslog-ng[2289]: last message repeated 2 times
Feb 16 22:28:46 pakanana kernel: EDAC MC0: CE page 0xa0000, offset 0x0, grain 536870912, syndrome 0x0, row 5, channel 0, label "": AMD762
Feb 16 22:28:47 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:54 pakanana syslog-ng[2289]: last message repeated 6 times
Feb 16 22:28:54 pakanana kernel: EDAC MC0: CE page 0x80000, offset 0x0, grain 536870912, syndrome 0x0, row 4, channel 0, label "": AMD762
Feb 16 22:28:55 pakanana kernel: EDAC MC0: CE page 0xc0000, offset 0x0, grain 536870912, syndrome 0x0, row 6, channel 0, label "": AMD762
Feb 16 22:28:56 pakanana kernel: EDAC MC0: CE page 0x80000, offset 0x0, grain 536870912, syndrome 0x0, row 4, channel 0, label "": AMD762


[Pakana]

Siis jos tuollainen määrä virheitä jäisi korjaamatta - eli ecc olisi biosista kytketty pois systeemi ei kauaa nikottelisi?
Kaikki pelaa ihan normaalisti myös silloin kun ecc on kytketty pois, voisi siis viitata siihen että emo ei oikein tule tuon ecc:n kanssa toimeen.
Noita virheitä se siis alkoi tuuppaamaan kun emolevy vaihtui.

[jmp]

Minkä merkkinen/mallinen emo kyseessä on?

Itselläni oli emo joka mukana tulleen manuaalin mukaan tuki ecc-muisteja,
mutta tosiasiassa se ei toimintoa tukenut (ja uudemmasta manuaalin versiosta kohta oli poistettu kokonaan).

[Pakana]

Asus A7M266-D, Muisteina Kingstonin Valueram DDR ECC REG palikoita 3x1GB ja 1x 512MB.

Muisteissa ei kyllä pitäisi vikaa olla, tai ainakin ne on Epoxin emolla toiminut vuosia kunnolla.

[jmp]

Olisikohan itselläni ollut tuo sama emo.

Lainaus sivulta http://www.ebabble.net/html/a7m266-d.html

With four DIMM slots there’s a lot of headroom for RAM. If you’re using standard DDR RAM, only the first two slots can be utilized.
If using registered DDR RAM all four slots can be filled, which is what I did with four PC2700 ECC Registered 512 MB DIMMs.
For some reason, no matter what type of ECC Registered DIMMs I used the BIOS always showed registered non-ECC RAM.
Switching brands and slots made no difference. My other machines recognized the RAM as ECC, but this board wouldn’t.

[Pakana]

Niin, kullä mulla tuo bios tunnistaa muistin ECC:ksi ja siellä on se vipukin josta ecc väännetään päälle.

[jmp]

Ainakin itselläni noita virhe/tila-viestejä tuli ecc:n takia /var/log/messages-tiedostoon aika paljon.
Kun muistit vaihtoi non-ecc-muisteihin niin järjestelmä toimi samalla tavalla, ilman noita ylimääräisiä viestejä messages-tiedostossa.

[Pakana]

Joo, sovitaan sitten että emo ei vaan hanksaa noita ecc muisteja. systeemui tosiaan toimi normaalisti vaikka ecc ei ollut päällä, jota se tuskin olisi tehnyt jos virheitä tosiaan noin tiiviisti tapahtuisi.