Ask Your Question
2

Fedora 20 graphics (?!) slow, no matter if gnome 3, gnome classic, KDE, Mate

asked 2014-01-29 08:34:37 -0500

hh-- gravatar image

updated 2014-01-30 11:26:07 -0500

mether gravatar image

Hi there,

[Solved - see bottom of the question for summary and answers to the question for details]

As a first note: This question might be related to https://ask.fedoraproject.org/en/question/37783/solved-installed-fedora-20-on-new-computer-and-it-is-very-slow-need-help/ but since the title of that one says "solved" although it was never explictly mentioned how I am reposting here. Please feel free to give me hints on how I can "link" this to the other one.

So now to my problem: I have recently upgraded from Fedora 17 to Fedora 20 doing a full fresh install (remove everything, install from scratch). Prior to the upgrade everything ran snappy. After the upgrade the system runs slow so that it is hard to really work with it. Let me define this slow a bit more:

  • When I e.g. switch from one window to another or switch desktops, having to wait 3-5 seconds is usual.
  • When I am writing an email with e.g. Thunderbird I can write faster than the words will be displayed in the mail (talking seconds here) If I have to remove characters or put my cursor at a different spot in the same mail, it takes up to 5 seconds to finally be there.
  • Skype used to be working easily, but now all voice transmission from the peer I talk to is getting to me in broken up pieces (the other side can understand me fine though). I am sure this is not to the bandwidth for my downstream or their upstream - since a) it used to work with Fedora 17, and b) (sorry to say that ;-) ) it works fine with both iOS and Windows 7.

First of all - I though this is maybe due to Gnome 3, but I tried, Gnome 3, Gnome classic, KDE, Mate and they are all more or less showing a similar behavior.

I am running a fairly decent HP laptop (EliteBook 2570p) with an 4-core Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz and 8 GB RAM, so I am pretty confident this problem is not due to not enough system resources. The CPU also does not show up to be too busy. As before I have a /boot a / and a swap partition (on the same disk). I use luks as before on / and swap. swap is 4GB as before (and also far from being full).

So I feel that this might be a graphic driver issue. Actually the only thing that I did different from last installs was leaving my laptop in a docking station with a 24 inch monitor attached when I did the install. No errros btw in the install itself.

I did all the recent system updates that were published but now I really don't have a clue why this problem is happening and/or how to track it down and I am hoping someone here can help. I am using my laptop for business purposes and this is the last try before I would probably have to roll ... (more)

edit retag flag offensive close merge delete

Comments

I had a similar problem The hard disk LED indicates much activity on the hard drive? How it works with an external monitor?

Gileade Oseiass gravatar imageGileade Oseiass ( 2014-01-29 09:01:20 -0500 )edit

Hi Gileade,

thanks for the quick response - indeed - when I switch workspaces or windows the hard disk LED goes on and it seems related to that. Well of course there are many programs running in the background so I cannot be absolutely sure but it least there is regular disk activity triggered by my switching it seems. This is with external monitor over the docking station. Is there any way I can provide reliable data for this to you (measure who is hitting the disk while I am switching ?!?). I will give it a try without the external monitor shortly. Thanks again

hh-- gravatar imagehh-- ( 2014-01-29 09:12:51 -0500 )edit

as an additional comment - it seems there is a relation between the smart test and the snappiness. Since it was hanging again, I just gave it a try to start another extensive smart test and (unless it is coincidence again) this gave me back the snappiness ?!? Obviously I cannot run smart tests all the time in a loop but it might give a hint to you or someone else as for what the root cause of this is. Just for completeness sake - I did not pull the laptop out of the docking station again before I started the new smart test. I btw use ext4 if that plays any role.

Any thoughts on what the root cause could be (or even how to potentially fix the issue)?

hh-- gravatar imagehh-- ( 2014-01-29 12:26:45 -0500 )edit

@hh-- please remove the huge logs from the comments. They are annoying. Use http://fpaste.org and post the link instead.

NickTux gravatar imageNickTux ( 2014-01-29 12:52:35 -0500 )edit

@NikTh Thanks for the hint and apologies for not doing it right directly. Will do for future comments. I just tried to edit the logs but there is an popup "You can only edit comments 10 minutes after they have posted or if it is the last comment". Any way to get past that?

hh-- gravatar imagehh-- ( 2014-01-29 13:50:25 -0500 )edit

2 Answers

Sort by ยป oldest newest most voted
1

answered 2014-01-29 17:15:16 -0500

NickTux gravatar image
SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x002f   100   100   050    Pre-fail  Always       -       0
  2 Throughput_Performance  0x0027   100   100   050    Pre-fail  Always       -       0
  3 Spin_Up_Time            0x0023   100   100   002    Pre-fail  Always       -       2254
  4 Start_Stop_Count        0x0032   100   100   000    Old_age   Always       -       763
  5 Reallocated_Sector_Ct   0x0033   092   092   010    Pre-fail  Always       -       182
  7 Seek_Error_Rate         0x002f   100   100   050    Pre-fail  Always       -       0
  8 Seek_Time_Performance   0x0025   100   100   050    Pre-fail  Offline      -       0
  9 Power_On_Hours          0x0032   082   082   000    Old_age   Always       -       7482
 10 Spin_Retry_Count        0x0033   115   100   030    Pre-fail  Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       763
183 Runtime_Bad_Block       0x0032   100   100   001    Old_age   Always       -       2
184 End-to-End_Error        0x0033   100   100   097    Pre-fail  Always       -       0
185 Unknown_Attribute       0x0032   100   100   001    Old_age   Always       -       65535
187 Reported_Uncorrect      0x0032   100   100   000    Old_age   Always       -       0
188 Command_Timeout         0x0032   100   100   000    Old_age   Always       -       0
189 High_Fly_Writes         0x003a   100   100   001    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0022   044   042   040    Old_age   Always       -       56 (Min/Max 28/58)
191 G-Sense_Error_Rate      0x0032   100   100   000    Old_age   Always       -       24
192 Power-Off_Retract_Count 0x0022   100   100   000    Old_age   Always       -       3932220
193 Load_Cycle_Count        0x0032   078   078   000    Old_age   Always       -       226361
196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       39
197 Current_Pending_Sector  0x0032   100   100   000    Old_age   Always       -       0
199 UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -       2

What we can figure out from above results is that the HDD needs to be replaced. What are the values that indicate such action ? From where we can see that this disk is damaged ?

  5 Reallocated_Sector_Ct   0x0033   092   092   010    Pre-fail  Always       -       182
  196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       39

What a reallocated sector is ? How this RAW_VALUE is a red alert for the health of the HDD ?

As we can read from Wikipedia

Reallocated_Sector_Ct

Count of reallocated sectors. When the hard drive finds a read/write/verification error, it marks that sector as "reallocated" and transfers data to a special reserved area (spare area). This process is also known as remapping, and reallocated sectors are called "remaps". The raw value normally represents a count of the bad sectors that have been found and remapped. Thus, the higher the attribute value, the more sectors the drive has had to reallocate. This allows a drive with bad sectors to continue operation; however, a drive which has had any reallocations at all is significantly more likely to fail in the near future.[3] While primarily used as a metric of the life expectancy of the drive, this number also affects performance. As the count of reallocated sectors increases, the read/write speed tends to become worse because the drive head is forced to seek to the reserved area whenever a remap is accessed. If sequential access speed is critical, the remapped sectors can be manually marked as bad blocks in the file system in order to prevent their use.

and

Reallocation Event Count

Count of remap ...

(more)
edit flag offensive delete link more

Comments

Thank you so much for this detailed answer and explanation. The only thing that I don't understand is - if this is critical and the disk is likely to fail - why in the world doesn't the SMART test issue an obvious warning ?

As far as I understand one of the purposes of SMART is to actually give explicit hints for when the disk is likely to fail. A PASSED indicates on the first sight and for users not dealing with disk counters regularly, that everything is OK with the disk.

Thanks again for your help Gileade and NikTh - you saved my future day without a hdd failure at a customer side!

hh-- gravatar imagehh-- ( 2014-01-30 08:37:46 -0500 )edit

Multiple tests exist. Long tests, short tests, I don't think that you ran them all, did you ? The most crucial thing here is what s.m.a.r.t has to say about your disk's health. Not the tests itself, so much. And s.m.a.r.t indicates that lot of reallocated sectors exist. This is red alert for disk's health. Nobody can tell you when the HDD will fail completely, but it's a common sense that when the disk "cracks" like this, someday will fail.

NickTux gravatar imageNickTux ( 2014-01-30 08:46:14 -0500 )edit

I did run the short test (below 1 minute) and the extended test (guess 20minutes). In any case the lesson learned is what you point out - it is not sufficient to just look at the SMART summary outcome only but the actual counters have to be looked at.

hh-- gravatar imagehh-- ( 2014-01-30 09:31:56 -0500 )edit
0

answered 2014-01-29 09:51:21 -0500

Gileade Oseiass gravatar image

updated 2014-01-30 08:57:39 -0500

Use the smartd command to test your hard drive.

# smartd or # smartctl -a /dev/sda

The smartd command without arguments will give a summary result of your hard disk

To understand the results of smartctl see the link below https://fedoraproject.org/wiki/Smartctl

edit flag offensive delete link more

Comments

@Gileade: Thank you very much for helping me tracking down the issue to the point that I could provide something helpful to be analyzed. Your both support is very appreciated!

One little thing - I know what you mean but there seems to be a little twist of characters in the command (samrtctl), maybe you can adjust that for future readers.

hh-- gravatar imagehh-- ( 2014-01-30 08:43:59 -0500 )edit

One hand washes the other thank you

Gileade Oseiass gravatar imageGileade Oseiass ( 2014-01-30 08:53:02 -0500 )edit

Question Tools

1 follower

Stats

Asked: 2014-01-29 08:34:37 -0500

Seen: 3,109 times

Last updated: Jan 30 '14