Following reports of broken updates, Apple has pulled the iOS 9.3.2 update for the latest-generation 9.7-inch iPad Pro, promising that a fix is in the works.
Owners of the new, smaller iPad Pro can no longer download the iOS 9.3.2 update from Apple. The issues do not affect other devices, like the 12.9-inch iPad Pro or iPhone models, and the iOS 9.3.2 update remains available for all other models.
Reports first began to surface earlier this week that the iOS 9.3.2 update was "bricking" some 9.7-inch iPad Pro models. It's unclear what caused the issues, as many users were able to successfully install the update without any problems.
Upon attempting to install the update, affected iPad Pro units displayed an "Error 56" message, which Apple identifies as a generic hardware effort. Early reports suggested there was no immediate workaround other than taking the tablet to an Apple Store for help or a replacement.
After pulling the update on Friday, Apple issued a statement to iMore stating it is working on a fix for the bug, which it claims affected "a small number of iPad units." The company pledged to deliver a new update "as quickly as possible."
iOS 9.3.2 launched on Monday, most notably restoring the ability to use Night Shift while in Low Power Mode. It also squashed a bug that caused sound quality issues when taking a call over a Bluetooth headset.
Apple also said iOS 9.3.2 fixes an issue where looking up dictionary definitions could fail, and it addresses a bug that prevented typing email addresses when using the Japanese Kana keyboard in Mail and Messages. The update also fixes an issue for VoiceOver users using Alex voice, where the device switches to a different voice to announce punctuation or paces.
Finally, Apple also said that iOS 9.3.2 fixes an issue that prevented MDM servers from installing Custom B2B apps.
4 Comments
this happened to me on the previous update. I just passed the 14 day return period when I did the update and it bricked out on me with the Error 56. I took it to Apple and even they had to scratch their heads. They did take care of me though - because the device just came out they didn't have replacements and since I had just passed the 14 day mark, they went to the back got me a shiny new one with the box and all. Kinda sucked that it's happening with this release but, because of what happened before, I didn't install the 9.3.2 straightway and I'm glad I waited.
I hope Apple squash this bug instead of doing this to the customer.
As far as it's statistical important... bought yesterday this iPad and made the update to 9.3.2. Everything ok.
I noticed that iOS 9.3.2 and iTunes 12.4.0.119 are significantly faster with Sync that any previous combinations I've seen. When I was syncing my various iDevices of various vintages I was amazed at how much faster each device was synching. When I hit my iPad Pro 12.9" I was sitting back ready to watch the magic speed boost taken to the pinnacle with Apple's fastest iOS device. But no - very soon into the sync sequence while it was downloading a medium sized iBook update to the iPad Pro it stopped downloading. iTunes became unresponsive and the iPad Pro showed the spinning arrow that it shows when data is in-flight but not yet committed. After waiting 10 minutes or so I tried to close iTunes to get it to abandon the Sync but it would not close. I had to Force Quit iTunes. When I reopened iTunes it couldn't find all of my iDevices. I rebooting and reset the PRAM and everything was restored to normal and I was able to Sync the iPad Pro.
This smells suspiciously like a synchronization/timing bug involving the writing and verifying *updates to the iOS mass storage. The iPad Pro is probably a bit faster than any other iOS devices and the embedded software engineers have a bug in their code that is being surfaced more frequently by the increased performance of the iPad Pro and it is more pronounced on the 9.7" device but still possible on other devices if the combination of events hits the timing window.
I've always been troubled with Apple's synchronization strategy with iTunes and iOS. I'd greatly prefer that the process be slower but failsafe rather than going for the best possible performance without transactional integrity. Seeing the iOS device still chewing on downloads (as evidenced by the spinning arrow) long after iTunes says it's "done" bothers me. You don't know whether data is still in flight or where it is in the update process. I've had a couple of instances of iOS devices having to be reformatted from scratch because I disconnected the device when iTunes said it was "done" but the iOS device was still churning. After those occurrences I will never unplug an iOS device after it's been downloaded until the iOS device no longer shows the thin spinning arrow.
I don't think Apple is unique in struggling with achieving 100% reliability with firmware/embedded SW updates to existing products. It's a challenging problem and it can be very expensive to provide redundancy based solutions. Users also expect high availability of their products and don't like waiting for updates to complete. They also expect all the newest bells and whistles on older products with limited processing and storage resources. All of these factors and others, such as the fact that most programmers struggle to write reliable multithreading and multiprocessing based applications and services leads to inevidable problems like the ones afflicting the iPad Pro 9.7 and iTunes, not to mention the majority of security related bugs in products that rely on software at any level. I hope the increased performance we're seeing with these newer updates are not indicative of programmers cutting corners, whether intentional or based on a naive understanding of the complexities of the problem domain. Fingers crossed.