Frequent session backups may decrease SSD lifetime

[LigH1L]

The ServeTheHome blog reports an issue with default settings regarding frequent writes to the user account which may be heavier than expected, thus reducing the lifetime of SSD's which are rather useful for frequent reads but sparse writes.

Firefox is eating your SSD – here is how to fix it

I don't have much experience with an SSD yet, so this issue may or may not be serious; would be interesting to know more. But better you know about it, maybe PaleMoon can be optimized in this regard somehow, or adapt to a present situation. The proposed fix is to increase this value considerably (but side effect will be decreased crash resilience): browser.sessionstore.interval

[adisib]

PM already has this set higher than Firefox. Firefox is set at 15,000, and PM at 60,000. I use 90,000 myself. I think the every 15 seconds Firefox uses is just ridiculous.

[Lunix]

Question - if I were to turn off the Session Restore feature altogether, would those writes completely stop or would I still have to reset the interval value as well? (I don't need Session Restore and am fine with just going back to a visited URL manually in the browser history.)

[ron_1]

Lunix wrote:
Question - if I were to turn off the Session Restore feature altogether, would those writes completely stop or would I still have to reset the interval value as well?

How do you turn it off?

[Lunix]

Lunix wrote:
Question - if I were to turn off the Session Restore feature altogether, would those writes completely stop or would I still have to reset the interval value as well?

How do you turn it off?

According to Mozilla, set browser.sessionstore.resume_from_crash to false.

There are several other browser.sessionstore settings - do I need to change anything else?

[adisib]

Lunix wrote:
Question - if I were to turn off the Session Restore feature altogether, would those writes completely stop or would I still have to reset the interval value as well?

How do you turn it off?

browser.sessionstore.max_tabs_undo = 0
browser.sessionstore.max_windows_undo = 0

[dark_moon]

Can't recommend that settings.
If your browser crash you lost all tabs.

Pale Moon doesn't make crazy stuff to your HDD/ SSD. So that settings are not necessary.

[Moonchild]

You're making a mountain out of a molehill.

Writing a few dozens or hundreds of KB once a minute will hardly wear down your SSD (unless it's really crappy and should actually be a USB stick ).
I found Mozilla's choice of interval poor as well (it previously was every 10 seconds!) and bumped it to a minute a long time ago, before SSDs were really a thing or commonplace.

Sure, it may (and probably will) decrease SSD lifetime, but the question is "by how much?" - ANY write, no matter how small, will decrease the lifetime of any SSD because the chips simply only last so long as far as rewriting the flash memory contents are concerned. I've done some checking previously as far as using disk cache on an SSD is concerned and that was insignificant in terms of expected life of an SSD (and that would have a lot more impact than any sessionstore writing would have).

You want to buy an SSD, but not have it wear down? That's like buying a sports car to just have it sit in front of your house and never use it.

(As the article even says, too, their much larger measurements went into other files in the profile -- and I also think measuring raw I/O is not a good representation of what is physically written to disk).

[LigH1L]

I guess having a swap file on SSD would give a similar result if you run so many applications that it has to be used.

[Moonchild]

I guess having a swap file on SSD would give a similar result if you run so many applications that it has to be used.

On the upside, your swap will be blazingly fast

[Trippynet]

It's an interesting one, so I decided to do some investigation into it.

I have a 256GB Samsung 840 Pro SSD as the main drive in my PC. I've had it for almost exactly 3 years (bought it in September 2013), and my PC is in daily use for about 6 hours a day (more on weekends) with Pale Moon running almost constantly (and with at least 30 tabs open). Now, my drive C has my OS, all my programs, and a moderate amount of my data on it (music, TV and games are on a separate mechanical hard drive, but documents and pictures are on drive C).

According to the Samsung Magician tool, I have written a total of 3.87TB to my drive in three years (about 3.5GB a day on average).

Looking around at the expected life of my drive, if I'd written 10GB a day, it'd have an expected lifespan of about 60 years (source). And I'm well under this.

Similarly, another site tested my drive to destruction and it managed about 2.4PB (that's 2,400TB) of writes before it failed (source), and over 600TB of writes before it had to start re-allocating sectors. It's also interesting to note that the worst SSD tested by this second site managed about 800TB of writes before it failed.

Hence with 3.87TB of data written in 3 years, I've barely scratched the surface of lifespan with my SSD - it should be obsolete long, long before it fails due to excessive use. And even with Firefox writing about 10GB of data a day to this guy's SSD (from the original article), it would also barely scratch the surface of SSD durability (see the 60 year estimated lifespan above).

In other words, this whole thing about browsers causing excessive writes and therefore killing SSDs prematurely is kind of making a colossal mountain out of an insignificant molehill.

[LigH1L]

I wonder if software like VRAMDir would be a useful solution for session backups. They may be useless after a reboot anyway, but will persist if only the browser crashes but the OS continues running. But it seems to be hopelessly obsolete, made in times of Win95.

[Octopuss]

Oh please, not SSDs and wear again.
Here, have a read: http://techreport.com/review/27909/the- ... e-all-dead

[GMforker]

Oh please, not SSDs and wear again.
Here, have a read: http://techreport.com/review/27909/the- ... e-all-dead

See also http://diit.cz/tagy/ssd-deep-hell (in Czech language).

But... It is necessary to note the following:

The Corsair, Intel, and Kingston SSDs all issued SMART warnings before their deaths, giving users plenty of time to preserve their data. The HyperX's warnings ended up being particularly premature, but that's better than no warning at all. Samsung's own software pronounced the 840 Series and 840 Pro to be in good health before their respective deaths. Worryingly, the 840 Series' uncorrectable errors didn't change that cheery assessment.
If you write a lot of data, keep an eye out for warning messages, because SSDs don't always fail gracefully. Among the ones we tested, only the Intel 335 Series and first HyperX remained accessible at the end. Even those bricked themselves after a reboot. The others were immediately unresponsive, possibly because they were overwhelmed by incoming writes before attempted resuscitation.

Of course, the same also applies to hard drive (comparatively).

Off-topic:
Ad viewtopic.php?f=56&t=13385. I'm sorry.

[bawldiggle]

.

Oh please, not SSDs and wear again.
Here, have a read: http://techreport.com/review/27909/the- ... e-all-dead

Thank you for the link. Wife gives her PC a flogging all day every day (video editing), and I wondered where is the failure point ?
I presume there must be software to assess SSD health and life expectancy ?

[back2themoon]

I presume there must be software to assess SSD health and life expectancy ?

The SSD manufacturer should provide that, check their website. There's also specialist software like the excellent Hard Disk Sentinel.

[LigH1L]

Some more or less related and useful tools can also be found on http://hddguru.com/

[bawldiggle]

@ back2themoon
@ LigH1L

Thank you for the links and tips