| 1 |
On Tue, Apr 25, 2017 at 08:51:22AM +0100, Peter Humphrey wrote: |
| 2 |
|
| 3 |
> > >> Quite so, but it was an internal compiler error, so that may not help. |
| 4 |
> > >> I'll remerge it when the machine's quiet and see. |
| 5 |
> > > |
| 6 |
> > > It did help. It's emerged okay. |
| 7 |
> > |
| 8 |
> > In my experience, internal GCC errors that show up intermittently |
| 9 |
> > and/or only under heavly load usually means failing RAM (or more |
| 10 |
> > rarely, some other hardware problem: something bad on the PCI bus, |
| 11 |
> > failing swap parition, etc.). |
| 12 |
> > |
| 13 |
> > I'd run memtest86 overnight, if I were you... |
| 14 |
> |
| 15 |
> Oo-er! I hope it's not that: this box is only just out of guarantee. |
| 16 |
|
| 17 |
I can confirm this. End of last year I experienced more and more internal |
| 18 |
errors. Turned out that I indeed had a faulty RAM module. The fault was |
| 19 |
shown by memtest86 almost instantly. I then rebuilt my system from scratch, |
| 20 |
because I coulnd't know what parts of my system had already been affected. |
| 21 |
|
| 22 |
Thankfully I have two of those 16 gig kits installed, so I can still run |
| 23 |
comfortably on half the fumes. |
| 24 |
-- |
| 25 |
Gruß | Greetings | Qapla’ |
| 26 |
Please do not share anything from, with or about me with any social network. |
| 27 |
|
| 28 |
One cannot be too careful when choosing one’s enemies. |
Archives