原文: UPDATE-2: After Christopher K.'s feedback I conducted further tests to isolate that "cyclical head activity" situation.
I first unplugged CAT6 cable to my NAS (Synology DiskStation DS213air). Wi-fi is already disabled. So NAS become absolutely stand-alone, away from all external interac
tion. When you unplug the ethernet cable, it goes thru a busy few minutes with disk activitiy. Then after it "calmed down", I noticed that such semi-cyclical head activity is still going on. At that point the only possible intervention to drives can obviously be coming from NAS operating system (DSM 6.1.3-15152 Update 3) and/or these 3 "packages" running on that NAS OS:
-Universal Search (can't disable)
-File Station (can't disable)
-Media Server (can uninstall)
So in theory either NAS OS or one of these packages can be "polling" the drives cyclically. Otherwise, such cyclical activity must be coming from drive itself. I checked a bit more online for that ST8000AS0002 drive. Looks like Seagate part no is 1NA17Z and firmware version of my drives is: RT17 . Seagate website informs that; "No Newer Firmware Available / A field update is not available" (I guess they don't want to give users the ability to change firmware - potentially discovering built-in "NSA code"! (as discovered by Kaspersky few years ago)).
Since user Christopher K. reported silent NAS operation of his same drives; the chances are that they might have never firmware. Or it could be purely because of NAS OS.
I update both OS and packages on my Synology NAS as they become available. So it might be theoretically possible that one such update (which coincided with me installing these new Seagate drives) could be the reason behind such cyclical activity. But I noticed this nuisance immediately after upgrading to these new Seagate drives. So I still reasonably think that it's being caused by the drive itself (whether SMR structure or firmware level).
On my test, next I plugged back CAT6 cable to my NAS. So now it's hardwired to my router (Netgear Nighthawk R6700); another potential source of external polling to the drives. Again after few more minutes of "calming down"; already existing activity level of drives didn't change. So the router is not guilty.
At that point my hard-wired desktop PCs are off. I turned on one of my Win10 PCs (running Kaspersky Internet Suite); still same activity level on those Seagate drives.
So my conclusion is; I still think that these drives (at least the ones I have - with potentially earlier firmware) are not able to "sit quite" in my given NAS. The only absolute proof will be when I upgrade them in the future (it'll be a non-SMR drive). If the new drives will be as quite as my earlier (non-SMR) Seagate drives; then I'll know that the reason was those drives after all...
--------------------------------------------------------------------------------------------------------------
UPDATE: In my NAS; drive heads seek in irregular intervals (about every 5-15 seconds), even there is no user read/write activity going on. Drives might be refreshing their buffers or something. Previous Seagate models (non-SMR type) were not doing that.
Other than that extremely audible nuisance, this non-stop mechanical activity might shorten life-span of drives. I'll update if they fail prematurely. I suggest that you avoid this (and any) drive that uses SMR technology.
--------------------------------------------------------------------------------------------------------------
Inside the unit is a Seagate Archive HDD v2 ST8000AS0002 8TB 5900 RPM 128MB Cache SATA 6.0Gb/s 3.5" Internal Hard Drive.
I needed the internal drive. This external unit was cheaper than buying the bare drive. Plus I have an extra USB3-to-SATA board with power adapter.
This drive uses SMR technology; so it's supposed to be slow (not to be used as system drive - for archive purposes only).
I replaced one of my aged NAS drives (also a Seagate 4TB Barracuda ST4000DM000 which reliably served me about 4 years - 24 hrs) with this new ST8000AS0002. Both drives have 5900 rpm speed. New one uses SMR technology, old one does not. New one runs at 100 F temperature (measured by my NAS) - old one at 90 F side-by side in NAS enclosure (fan cooled).
I use my NAS strictly with ethernet cable (although it has wi-fi option; it's a joke and I disabled it after trying it once). So with old drives (in parallel writing RAID mode) I was getting 75MB/sec transfer speed. Surprisingly with one new drive (other 4TB still keeps running after 4 years) I still get the same speed. So speed of this 8TB SMR drive is totally acceptable to me. Much more than archiving speed, I can consistently stream HD movie with that 75MB/sec hard-wired ethernet connection.
For those who are thinking about doing the same; opening the case is pretty much destructive (read: your warranty for whole unit is void). Interestingly; I checked warranty status for bare internal drive : Seagate web site informs that "the unit was sold as a system component. Please contact your place of purchase for service"... So after all; directly buying an internal drive (and paying a bit more) makes better sense for warranty purposes.
I'll update later on about longevity of the new drive.
显示更多>>
译文: 更新-2:在克里斯托弗K的反馈后,我进行了进一步的测试,以解决“周期性端口活动”的情况。
我首先将CAT6电缆拔出至,插到我的NAS上(Synology DiskStation DS213air)。无线网络已经被禁用。所以NAS变成了完全独立的,远离了所有的外部交互。当你拔下以太网电缆时,它的磁盘活动会开始忙绿的几分钟。然后在“平静下来”之后,我注意到这种半周期的端口活动还在继续。此时,唯一可能的驱动干预显然可能来自NAS操作系统(DSM 6.1.3-15152 Update 3)或运行在该NAS OS上的这3个“软件包”:
- 通用搜索(无法禁用)
文件站(不能禁用)
- 媒体服务器(可以卸载)
所以从理论上讲,NAS OS或其中的一个软件包可以循环地“访问”驱动器。否则,这样的周期性活动一定是来自驱动器本身。我在网上查了一下这个ST8000AS0002驱动器。看起来希捷部件号是1NA17Z,我的驱动器的固件版本是:RT17。希捷网站通知说; “没有新的固件可用、更新不可用”(我想他们不希望有给用户更改固件的能力 这样可能会发现内置的“NSA代码”!(几年前由卡巴斯基发现的)) 。
由于用户克里斯托弗K说了他的同一驱动器的无声NAS操作;有可能是从来没有固件。或者它可能纯粹是NAS操作系统导致的。
当我的Synology NAS可用时,我会更新操作系统和软件包。因此理论上可能的是,这样的更新(与我一起安装这些新的希捷驱动器时)可能是此类周期性活动背后的原因。但升级到这些新的希捷硬盘后,我立即注意到了这个麻烦的问题。所以我还是有理由认为这是由驱动器本身引起的(无论是SMR结构还是固件级别)。
在我的测试中,接下来我将CAT6电缆插回到我的NAS上。所以现在它强行插入到了我的路由器中(Netgear Nighthawk R6700);驱动器的另一个外部访问的潜在来源。几分钟之后再次“平静下来”;已经存在的驱动器活动水平没有改变。所以不关路由器的事。
那时我的连线台式电脑关闭了。我打开了我的一台Win10电脑(运行卡巴斯基互联网套件)。这些希捷驱动器的活动水平仍然相同。
所以我的结论是;我仍然认为这些驱动器(至少是我拥有的驱动器 - 具有更早时期的固件)不能用在我的给定的NAS上。唯一的绝对证明是我将来得升级它(这将是一个非SMR驱动器)。如果新驱动器和我以前的(非SMR)希捷驱动器一样,那么我就知道原因是因为这些驱动器...
-------------------------------------------------- -------------------------------------------------- ----------
更新:在我的NAS上;驱动器磁头以不规则的间隔寻找(大约每5-15秒),即使没有用户读/写活动,正它也会进行。驱动器可能会刷新缓冲区或其他东西。以前的希捷模型(非SMR类型)没有这样做。
除此之外,这种不间断的机械活动可能会缩短驱动器的使用寿命。我会更新评论,如果他们很快就坏了的话。我建议你避免使用SMR技术的这个(和任何)驱动器。
-------------------------------------------------- -------------------------------------------------- ----------
内置单元是一个希捷存档硬盘v2 ST8000AS0002 8TB 5900转128MB缓存的SATA 6.0Gb / s 3.5“内置硬盘驱动器。
我需要内置驱动器。这个内置单位比买光驱要便宜。另外我还有一个带电源适配器的USB3-to-SATA板。
该驱动器使用SMR技术;所以它运行速度应该很慢(不能用作系统驱动器 - 仅用于归档目的)。
我用一台新的ST8000AS0002替换了一个老旧的NAS驱动器(也就是希捷4TB Barracuda ST4000DM000,它让我很安心的每天使用,用了大约4年)。两个驱动器都有5900转的转速。新的使用了SMR技术,老的没有。新的一个在100F温度下运行(由我的NAS测量出来的) - 在90°F并排的NAS机箱(有散热器冷却)。
我使用我的NAS的时候,严格使用以太网电缆(虽然它有Wi-Fi选项,但是这是一个笑话,我试了一次后就禁用了它)。 所以用了旧的驱动器(并行写RAID模式),我得到了75MB/秒的传输速度。 令人惊讶的是,一个新的驱动器(其他4TB仍然保持可以运行4年),我仍然有同样的速度。 所以这8TB SMR驱动器的速度是完全可以接受的。 除了存档速度之外,我还可以用75MB /秒的有线以太网连接持续流式传输高清电影。
对于那些正在考虑这样做的人来说, 打开机箱是非常具有破坏性的(阅读:您的整个机器的保修是无效的)。有趣的是, 我检查了内置驱动器的保修状态:希捷网站通知:“该设备作为系统组件出售,请联系您的购买地点以便进行维修”。 直接购买一个内部驱动器(并多付一点钱)对于保修来说更有意义。
稍后我会更新有关新驱动器的使用寿命的事情。