| 1 |
On Mon, Jun 18 2018, Mick wrote: |
| 2 |
|
| 3 |
> On Monday, 18 June 2018 21:28:38 BST allan gottlieb wrote: |
| 4 |
>> On Mon, Jun 18 2018, Mick wrote: |
| 5 |
>> > Hi Allan, |
| 6 |
>> > |
| 7 |
>> > On Monday, 18 June 2018 18:52:31 BST allan gottlieb wrote: |
| 8 |
>> >> On Sun, Jun 17 2018, Mick wrote: |
| 9 |
>> >> > On Sunday, 17 June 2018 22:04:50 BST allan gottlieb wrote: |
| 10 |
>> >> >> (I have been offline for a few months and apologize if this has been |
| 11 |
>> >> >> covered.) |
| 12 |
>> >> >> |
| 13 |
>> >> >> I had not synced for about 3 months (fear of new, incompatible |
| 14 |
>> >> >> gnucash) |
| 15 |
>> >> >> but have now done so. Unsurprisingly my normal update world shows |
| 16 |
>> >> >> many |
| 17 |
>> >> >> entries and also unsurprising is a blocker (slot conflict). |
| 18 |
>> > |
| 19 |
>> > As a first step I suggest you go through the 'eselect news read new'. |
| 20 |
>> > |
| 21 |
>> > There was a profile change sometime late last year. The profile change |
| 22 |
>> > enews is important and you should deal with it first. It is titled: |
| 23 |
>> > |
| 24 |
>> > 'New 17.0 profiles in the Gentoo repository' |
| 25 |
>> > |
| 26 |
>> > and it involves updating gcc as part of it and perhaps changing your |
| 27 |
>> > profile if it is due to be deprecated. |
| 28 |
>> |
| 29 |
>> I had done the upgrade to 17.0 months ago. |
| 30 |
>> 17.1 is unstable so I believe 17.0 is the right profile for me. |
| 31 |
>> |
| 32 |
>> > Once you've been through all this give it another spin, but use |
| 33 |
>> > --backtrack=99 to see if portage resolves the conflicts. |
| 34 |
>> |
| 35 |
>> backtrack=99 didn't change the situation, but adding in addition |
| 36 |
>> --autounmask-backtrack=y did! |
| 37 |
> |
| 38 |
> Interesting! What did it autounmask? |
| 39 |
|
| 40 |
That is not what the option does. It just prevents backtracking from |
| 41 |
stopping earlier when it finds that config changes are needed. |
| 42 |
With the full backtrack=99 all the serious problems were resolved, I |
| 43 |
just needed to add a use setting, specifically grub[mount]. |
| 44 |
|
| 45 |
>> > I can't see a [B] in the list you provided, all are small blocks [b] which |
| 46 |
>> > portage will deal with on its own. |
| 47 |
>> |
| 48 |
>> The list I provided was for my update world. That has little b's. The |
| 49 |
>> only blockage I have is the slot conflict I mentioned originally. The |
| 50 |
>> big B's happened when I tried |
| 51 |
>> |
| 52 |
>> emerge -1pDv media-video/ffmpeg |
| 53 |
> |
| 54 |
> Well, media-libs/x264-0.0.20130506 is no longer in the tree. media-video/ |
| 55 |
> ffmpeg-3.3.6 does not require it and works fine with media-libs/ |
| 56 |
> x264-0.0.20160712. This is why I suggested that manual removal and update |
| 57 |
> earlier on. |
| 58 |
> |
| 59 |
> Unless something else is blocking ffmpeg (you've got a long update list there |
| 60 |
> which might influence it) there shouldn't be a problem with my suggested |
| 61 |
> workaround. Nevertheless, getting portage to pontificate and resolve the |
| 62 |
> dependency graph is invariably a safer option. ;-) |
| 63 |
|
| 64 |
By just adding --backtrack=99 --autounmask-backtrack=y |
| 65 |
portage was happy (and so am I). |
| 66 |
|
| 67 |
Indeed, considering I hadn't synced for 3 months, I consider myself |
| 68 |
lucky and will try to do at least weekly emerge --update @world's |
| 69 |
|
| 70 |
thanks, |
| 71 |
allan |
Archives